{"index":{"byClaim":[["rsp-v2-1-rd-thresholds",["documents:frontier-safety-frameworks"]],["prov-rsp-v3-4-competitor-lead-delay",["documents:frontier-safety-frameworks"]],["cs-anthropic-ltbt-removes-directors",["control:anthropic"]],["prov-wh-2023-07-public-reporting",["commitments:external-commitments","guide:frontier-frameworks-and-law"]],["cm-seoul-xai",["commitments:external-commitments"]],["ts-b6-vinge-1983-context",["propositions:superintelligence"]],["yk-y08-sl4-2000-07-09-yudkowsky-0007-0017-b",["propositions:yudkowsky-arguments"]],["yk-y08-sl4-2006-06-06-yudkowsky-0606-15174-c",["propositions:yudkowsky-arguments"]],["yk-y13-sl4-2002-07-02-yudkowsky-0207-4667-b",["propositions:yudkowsky-arguments"]],["ec-g09-erdil-besiroglu-2023-explosive-growth-review-a",["propositions:economics-of-machine-intelligence"]],["ec-m01-leontief-duchin-1984-impacts-of-automation-b",["propositions:economics-of-machine-intelligence"]],["ec-m09-agrawal-gans-goldfarb-2018-economic-policy-for-ai-b",["propositions:economics-of-machine-intelligence"]],["cmp-preparedness-beta-2023-d3-safeguards",["compare:frontier-safety-frameworks"]],["cmp-rsp-v2-1-d8-process",["compare:frontier-safety-frameworks"]],["fl-eu-aia-art52-5-reassessment",["guide:frontier-frameworks-and-law"]],["fl-uk-cohere-mou-not-binding",["guide:frontier-frameworks-and-law"]],["rd-r06-leike-2022-optimistic-alignment-approach",["propositions:risk-debate"]],["rd-r07-hawking-russell-tegmark-wilczek-2014-independent",["propositions:risk-debate"]],["rd-r12-nature-2023-editorial-stop-talking",["propositions:risk-debate"]],["rd-r14-bender-koller-2020-climbing-nlu",["propositions:risk-debate"]]],"marks":[["diebold-1955-automation-condensation",[{"collectionId":"economics-of-machine-intelligence","propositionId":"automation-lowers-labour-income","claimId":"ec-m06a-diebold-1955-automation-condensation-a","text":"Earliest located in tier 2 of Economics of machine intelligence for M06a Automation lowers the labour share, wages or employment"},{"collectionId":"economics-of-machine-intelligence","propositionId":"compensating-policy-for-ai","claimId":"ec-m06b-diebold-1955-automation-condensation-a","text":"Earliest located in tier 2 of Economics of machine intelligence for M06b Compensating policy for AI’s distributional effects"}]],["extropians-1998-09-07-yudkowsky-2209",[{"collectionId":"superintelligence","propositionId":"seed-ai","claimId":"ts-a2-yudkowsky-1998-09-07-2209","text":"Earliest located in Propositions about superintelligence, 1863–2022 for A2 Seed AI"}]],["ai-positive-negative-factor-global-risk",[{"collectionId":"yudkowsky-arguments","propositionId":"agi-decades-away","claimId":"yk-y13-ai-positive-negative-factor-global-risk-c","text":"Earliest located text contesting Y13 AGI is decades away, so the problem can wait in Eliezer Yudkowsky’s arguments on common claims about AI, 1996–2026"}]]],"chronicle":{"years":[{"year":1863,"topics":{"ideas":5},"stated":0},{"year":1865,"topics":{"ideas":1},"stated":0},{"year":1872,"topics":{"ideas":7},"stated":0},{"year":1943,"topics":{"ideas":1},"stated":0},{"year":1947,"topics":{"ideas":2},"stated":0},{"year":1950,"topics":{"ideas":7},"stated":0},{"year":1951,"topics":{"ideas":5},"stated":0},{"year":1952,"topics":{"ideas":2},"stated":0},{"year":1955,"topics":{"ideas":31},"stated":0},{"year":1956,"topics":{"ideas":1},"stated":0},{"year":1957,"topics":{"ideas":1},"stated":0},{"year":1958,"topics":{"ideas":3},"stated":0},{"year":1959,"topics":{"ideas":6},"stated":0},{"year":1960,"topics":{"ideas":11},"stated":0},{"year":1961,"topics":{"ideas":10},"stated":0},{"year":1965,"topics":{"ideas":8},"stated":0},{"year":1966,"topics":{"ideas":20},"stated":0},{"year":1968,"topics":{"ideas":5},"stated":0},{"year":1969,"topics":{"ideas":2},"stated":0},{"year":1971,"topics":{"ideas":4},"stated":0},{"year":1972,"topics":{"ideas":1},"stated":0},{"year":1973,"topics":{"ideas":2},"stated":0},{"year":1974,"topics":{"ideas":2},"stated":0},{"year":1975,"topics":{"ideas":1},"stated":0},{"year":1976,"topics":{"ideas":1},"stated":0},{"year":1977,"topics":{"people":2,"ideas":8},"stated":0},{"year":1979,"topics":{"ideas":4},"stated":0},{"year":1980,"topics":{"ideas":2},"stated":0},{"year":1981,"topics":{"people":2},"stated":0},{"year":1982,"topics":{"ideas":4},"stated":0},{"year":1983,"topics":{"ideas":4},"stated":0},{"year":1984,"topics":{"ideas":21},"stated":0},{"year":1985,"topics":{"ideas":9},"stated":0},{"year":1986,"topics":{"ideas":18},"stated":0},{"year":1987,"topics":{"ideas":1},"stated":0},{"year":1988,"topics":{"ideas":6},"stated":0},{"year":1989,"topics":{"ideas":8},"stated":0},{"year":1990,"topics":{"ideas":5},"stated":0},{"year":1991,"topics":{"ideas":11},"stated":0},{"year":1992,"topics":{"ideas":9},"stated":0},{"year":1993,"topics":{"ideas":14,"people":3},"stated":0},{"year":1994,"topics":{"ideas":14},"stated":0},{"year":1995,"topics":{"ideas":10,"people":1},"stated":0},{"year":1996,"topics":{"ideas":71},"stated":0},{"year":1997,"topics":{"ideas":12},"stated":0},{"year":1998,"topics":{"ideas":49},"stated":0},{"year":1999,"topics":{"ideas":27,"people":1,"institutions":1},"stated":0},{"year":2000,"topics":{"ideas":62},"stated":0},{"year":2001,"topics":{"ideas":39},"stated":0},{"year":2002,"topics":{"ideas":30},"stated":0},{"year":2003,"topics":{"ideas":8},"stated":0},{"year":2004,"topics":{"ideas":12,"people":3},"stated":0},{"year":2005,"topics":{"ideas":10,"people":1},"stated":0},{"year":2006,"topics":{"ideas":26,"people":1},"stated":0},{"year":2007,"topics":{"ideas":20},"stated":0},{"year":2008,"topics":{"ideas":57,"people":3,"institutions":1},"stated":0},{"year":2009,"topics":{"ideas":10,"people":1},"stated":0},{"year":2010,"topics":{"people":3,"ideas":11,"institutions":1,"governance":4},"stated":0},{"year":2011,"topics":{"people":3,"ideas":7,"institutions":2},"stated":0},{"year":2012,"topics":{"people":4,"ideas":17,"institutions":2},"stated":0},{"year":2013,"topics":{"ideas":46,"people":4,"institutions":2,"capital":1},"stated":0},{"year":2014,"topics":{"institutions":2,"ideas":54,"people":5},"stated":0},{"year":2015,"topics":{"people":22,"ideas":122,"institutions":3,"capital":1},"stated":0},{"year":2016,"topics":{"ideas":88,"people":2},"stated":0},{"year":2017,"topics":{"people":16,"institutions":2,"ideas":128,"governance":2},"stated":0},{"year":2018,"topics":{"people":8,"ideas":158,"institutions":8,"governance":1},"stated":0},{"year":2019,"topics":{"capital":1,"compute":1,"people":10,"governance":19,"ideas":175,"institutions":4},"stated":0},{"year":2020,"topics":{"people":6,"ideas":98,"institutions":1,"governance":2},"stated":0},{"year":2021,"topics":{"people":27,"ideas":177,"compute":1,"institutions":23,"governance":4,"capital":6},"stated":0},{"year":2022,"topics":{"people":15,"ideas":217,"institutions":6,"governance":3,"compute":2,"capital":1},"stated":0},{"year":2023,"topics":{"capital":5,"compute":7,"institutions":23,"people":51,"governance":179,"ideas":389},"stated":0},{"year":2024,"topics":{"capital":12,"compute":19,"governance":282,"people":80,"ideas":178,"institutions":28},"stated":1},{"year":2025,"topics":{"governance":547,"compute":69,"capital":34,"people":90,"institutions":18,"ideas":178},"stated":7},{"year":2026,"topics":{"governance":442,"capital":8,"compute":32,"institutions":21,"ideas":91,"people":107},"stated":19},{"year":2027,"topics":{},"stated":9},{"year":2028,"topics":{},"stated":6},{"year":2029,"topics":{},"stated":3},{"year":2030,"topics":{},"stated":3},{"year":2032,"topics":{},"stated":2},{"year":2035,"topics":{},"stated":2},{"year":2039,"topics":{},"stated":1},{"year":2042,"topics":{},"stated":1},{"year":2050,"topics":{},"stated":1}],"scheduled":[],"statedBy":[[2028,[["cpo-pw-google-constellation-890mw","compute"]]]]},"collections":[{"target":{"kind":"propositions","id":"superintelligence"},"title":"Propositions about superintelligence, 1863–2022","claimIds":["ts-b6-vinge-1983-context"],"records":[]},{"target":{"kind":"propositions","id":"risk-debate"},"title":"Arguments about catastrophic risk from advanced AI","claimIds":["rd-r06-leike-2022-optimistic-alignment-approach","rd-r07-hawking-russell-tegmark-wilczek-2014-independent","rd-r12-nature-2023-editorial-stop-talking","rd-r14-bender-koller-2020-climbing-nlu"],"records":[]},{"target":{"kind":"propositions","id":"yudkowsky-arguments"},"title":"Eliezer Yudkowsky’s arguments on common claims about AI, 1996–2026","claimIds":["yk-y08-sl4-2000-07-09-yudkowsky-0007-0017-b","yk-y08-sl4-2006-06-06-yudkowsky-0606-15174-c","yk-y13-sl4-2002-07-02-yudkowsky-0207-4667-b"],"records":[]},{"target":{"kind":"propositions","id":"economics-of-machine-intelligence"},"title":"Economics of machine intelligence","claimIds":["ec-g09-erdil-besiroglu-2023-explosive-growth-review-a","ec-m01-leontief-duchin-1984-impacts-of-automation-b","ec-m09-agrawal-gans-goldfarb-2018-economic-policy-for-ai-b"],"records":[]},{"target":{"kind":"propositions","id":"orbital-compute-arguments"},"title":"Arguments about compute in orbit","claimIds":[],"records":[]},{"target":{"kind":"propositions","id":"ai-bio-arguments"},"title":"Arguments about AI and biological risk","claimIds":[],"records":[]},{"target":{"kind":"propositions","id":"ai-in-science"},"title":"AI in science","claimIds":[],"records":[]},{"target":{"kind":"documents"},"title":"Frontier safety frameworks","claimIds":["rsp-v2-1-rd-thresholds","prov-rsp-v3-4-competitor-lead-delay"]},{"target":{"kind":"compare","id":"frontier-safety-frameworks"},"title":"Compare frameworks","claimIds":["cmp-preparedness-beta-2023-d3-safeguards","cmp-rsp-v2-1-d8-process"]},{"target":{"kind":"control","id":"openai"},"title":"OpenAI’s control structure, 2015–2026","claimIds":[]},{"target":{"kind":"control","id":"anthropic"},"title":"Anthropic’s control structure, 2021–2026","claimIds":["cs-anthropic-ltbt-removes-directors"]},{"target":{"kind":"commitments"},"title":"External commitments and obligations","claimIds":["prov-wh-2023-07-public-reporting","cm-seoul-xai"]},{"target":{"kind":"question","id":"frameworks-to-obligations"},"title":"Frameworks, commitments and legal obligations, 2023–2026","claimIds":["prov-wh-2023-07-public-reporting","cm-seoul-xai"]},{"target":{"kind":"reading"},"title":"Founding and early leadership","claimIds":[]}]},"entities":[{"id":"openai","name":"OpenAI","type":"organisation"},{"id":"anthropic","name":"Anthropic","type":"organisation"},{"id":"xai","name":"xAI","type":"organisation"},{"id":"ilya-sutskever","name":"Ilya Sutskever","type":"person"},{"id":"dario-amodei","name":"Dario Amodei","type":"person"},{"id":"eliezer-yudkowsky","name":"Eliezer Yudkowsky","type":"person"},{"id":"rsp-2023","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v1.0","publisherId":"anthropic","publishedAt":"2023-09-19","effectiveAt":"2023-09-19","dateNote":"Initial public policy dated September 19, 2023; original entity and publication claim retained."}},{"id":"iem-2013","name":"Intelligence Explosion Microeconomics","type":"document"},{"id":"igor-babuschkin","name":"Igor Babuschkin","type":"person"},{"id":"google","name":"Google","type":"organisation"},{"id":"rsp-v2-0","name":"Anthropic Responsible Scaling Policy — v2.0","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v2.0","publisherId":"anthropic","effectiveAt":"2024-10-15","dateNote":"The stated effective date is retained; a separate publication day is not inferred."}},{"id":"rsp-v2-1","name":"Anthropic Responsible Scaling Policy — v2.1","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v2.1","publisherId":"anthropic","effectiveAt":"2025-03-31","dateNote":"The stated effective date is retained; a separate publication day is not inferred."}},{"id":"rsp-v2-2","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v2.2","publisherId":"anthropic","effectiveAt":"2025-05-14","dateNote":"The stated effective date is retained; a separate publication day is not inferred."}},{"id":"rsp-v3-0","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v3.0","publisherId":"anthropic","publishedAt":"2026-02-24","effectiveAt":"2026-02-24","dateNote":"The cover effective date and release announcement publication date both identify February 24, 2026."}},{"id":"rsp-v3-1","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v3.1","publisherId":"anthropic","effectiveAt":"2026-04-02","dateNote":"The stated effective date is retained; a separate publication day is not inferred."}},{"id":"rsp-v3-2","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v3.2","publisherId":"anthropic","effectiveAt":"2026-04-29","dateNote":"The stated effective date is retained; a separate publication day is not inferred."}},{"id":"rsp-v3-3","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v3.3","publisherId":"anthropic","effectiveAt":"2026-05-26","dateNote":"The stated effective date is retained; a separate publication day is not inferred."}},{"id":"rsp-v3-4","type":"document","documentVersion":{"family":"anthropic-rsp","label":"v3.4","publisherId":"anthropic","effectiveAt":"2026-07-08","dateNote":"The stated effective date is retained; a separate publication day is not inferred."},"name":"Anthropic Responsible Scaling Policy — v3.4"},{"id":"preparedness-beta-2023","name":"OpenAI Preparedness Framework — Beta","type":"document","documentVersion":{"family":"openai-preparedness","label":"Beta","publisherId":"openai","publishedAt":"2023-12-18","dateNote":"The sourced publication date does not establish an implementation or effective date."}},{"id":"preparedness-v2","type":"document","documentVersion":{"family":"openai-preparedness","label":"Version 2","publisherId":"openai","publishedAt":"2025-04-15","revisedAt":"2025-04-15","dateNote":"Cover last-updated date and dated publication announcement both identify April 15, 2025; no distinct effective date is asserted."}},{"id":"ssi","name":"Safe Superintelligence Inc.","type":"organisation"},{"id":"amplification-paper","name":"Supervising strong learners by amplifying weak experts","type":"document"},{"id":"reinforcement-learning","name":"Reinforcement learning","type":"concept"},{"id":"deepseek-r1-report","name":"DeepSeek-R1 research report","type":"document"},{"id":"zhipu","name":"Zhipu AI / 智谱","type":"organisation"},{"id":"zhang-peng","name":"Zhang Peng / 张鹏","type":"person"},{"id":"openai-inc","name":"OpenAI, Inc. (OpenAI Foundation)","type":"organisation"},{"id":"anthropic-ltbt","name":"Anthropic Long-Term Benefit Trust","type":"organisation"},{"id":"wh-voluntary-ai-commitments-2023-07","name":"Ensuring Safe, Secure, and Trustworthy AI: voluntary AI commitments (July 2023)","type":"document","documentVersion":{"family":"us-voluntary-ai-commitments","label":"July 2023 text","publisherId":"us-white-house","publishedAt":"2023-07-21","dateNote":"Date of the White House fact sheet that announced the commitments and linked this text; the PDF itself is undated."}},{"id":"wh-voluntary-ai-commitments-2023-09","type":"document","documentVersion":{"family":"us-voluntary-ai-commitments","label":"September 2023 text","publisherId":"us-white-house","publishedAt":"2023-09-12","dateNote":"Date of the White House fact sheet that announced the second round and linked this text; the PDF itself is undated."}},{"id":"frontier-ai-safety-commitments-2024","name":"Frontier AI Safety Commitments, AI Seoul Summit 2024","type":"document","documentVersion":{"family":"frontier-ai-safety-commitments","label":"GOV.UK text (May 2024; list updated February 2025)","publisherId":"dsit","publishedAt":"2024-05-21","revisedAt":"2025-02-07","dateNote":"First published on 21 May 2024; the GOV.UK change history records an update on 7 February 2025 that added organisations to the list. The commitment text itself is unchanged between the 2024-05-21 capture and the live page."}},{"id":"vinge-1983-first-word","name":"First Word","type":"document"},{"id":"extropians-1998-09-07-yudkowsky-2209","name":"Human AI to superhuman (Re: Max More)","type":"document"},{"id":"ai-positive-negative-factor-global-risk","name":"Artificial Intelligence as a Positive and Negative Factor in Global Risk","type":"document"},{"id":"rlo-2019-deceptive-alignment","name":"Deceptive Alignment (Risks from Learned Optimization, part 4)","type":"document"},{"id":"control-or-purpose-specification","name":"Control or purpose specification","type":"concept","proposition":{"code":"B6","criterion":"A claim that people may be unable to keep the ability to direct, correct or stop capable machines (the control half), or that a machine may pursue the purpose as specified rather than the purpose intended (the specification half). Loss of the power to stop machines through dependence counts for the control half; a bare prediction that machines will be superior does not.","terms":[]}},{"id":"warning-signs-in-time","name":"Clear warning signs will come before danger","type":"concept","proposition":{"code":"Y08","criterion":"An explicit claim that clear warning signs will appear before advanced AI becomes dangerous, on the premise that a recognisable, consensus-forming signal will come in time to act, that dangerous failure modes will first appear in weaker systems where they can be observed and fixed, or that a warning sign, once seen, will lead to effective action.","terms":[]}},{"id":"agi-decades-away","name":"AGI is decades away, so the problem can wait","type":"concept","proposition":{"code":"Y13","criterion":"An explicit claim that the risks of AGI can be left until later because AGI is decades away, on the premise that it can be known that AGI is decades away, or that long timelines mean there is no need to act now.","terms":[]}},{"id":"sl4-2000-07-09-yudkowsky-0007-0017","name":"Re: [SL4] Programmed morality","type":"document"},{"id":"sl4-2002-06-23-yudkowsky-0206-4219","name":"Re: Seed AI (was: How hard a Singularity?)","type":"document"},{"id":"sl4-2002-07-02-yudkowsky-0207-4667","name":"Re: Seed AI (was: How hard a Singularity?)","type":"document"},{"id":"eu-ai-act-2024","name":"Regulation (EU) 2024/1689 (Artificial Intelligence Act), as published in the Official Journal","type":"document","documentVersion":{"family":"eu-ai-act","label":"As published in the Official Journal (12 July 2024)","publisherId":"european-parliament-and-council","publishedAt":"2024-07-12","dateNote":"The Official Journal header gives 12.7.2024 and the act is dated 13 June 2024 (signature). Article 113 sets separate dates of entry into force and application by chapter; no single effective date is recorded for the edition. Regulation (EU) 2026/1744 of 8 July 2026 later amended it; the consolidated text was not inspected."}},{"id":"explosive-growth-estimates","name":"Estimates of explosive growth within a stated horizon","type":"concept","proposition":{"code":"G09","criterion":"The text states its own estimate of whether or when explosive growth (as the text defines it) occurs in the actual future, with machine intelligence or automation among the stated drivers. Each record gives the holder, the estimate as stated (a number or words), the horizon, the text's definition and the conditions. Records are not converted to a common threshold, combined or ranked; a low estimate and a high estimate both articulate this proposition.","terms":[],"withheld":[{"role":"earliest-located","reason":"Withheld on 2026-10-25 by maintainer decision. The earlier holder, Hanson’s “Economic Growth Given Machine Intelligence”, relays his separate 1998 projection. His “Long-Term Growth As A Sequence of Exponential Modes” (December 2000; first version September 1998) is an unchecked lead for the role.","recordedAt":"2026-10-25"}]}},{"id":"automation-as-ordinary-technical-change","name":"Computers and AI as ordinary technical change","type":"concept","proposition":{"code":"M01","criterion":"A tier-1 text dated before 2015 that discusses computers, automation or AI and analyses them with the tools used for technical change in general (factor-augmenting or skill-biased change, capital deepening), with no separate category for machine cognition.","terms":[]}},{"id":"recent-economics-engagement-with-ai","name":"Recent entry of machine intelligence into mainstream economics","type":"concept","proposition":{"code":"M09","criterion":"Not a criterion for a single passage. Assessed from the dated record: the earliest located tier-1 texts, within the searched scope, whose subject is AI, automation by AI or machine intelligence; plus explicit self-descriptions in tier-1 texts of the field as new or neglected. Tier-2 and outside texts are context.","terms":[]}},{"id":"erdil-besiroglu-2023-explosive-growth-review","name":"Explosive growth from AI automation: A review of the arguments","type":"document"},{"id":"agrawal-gans-goldfarb-2018-economic-policy-for-ai","name":"Economic Policy for Artificial Intelligence","type":"document"},{"id":"acemoglu-restrepo-2019-wrong-kind-of-ai","name":"The Wrong Kind of AI? Artificial Intelligence and the Future of Labor Demand","type":"document"},{"id":"diebold-1955-automation-condensation","name":"A Condensation of the Book Automation, in Automation and Technological Change: Hearings","type":"document"},{"id":"leontief-duchin-1984-impacts-of-automation","name":"The Impacts of Automation on Employment, 1963–2000. Final Report","type":"document"},{"id":"apple","name":"Apple","type":"organisation"},{"id":"baidu","name":"Baidu / 百度","type":"organisation"},{"id":"ben-mann","name":"Ben Mann","type":"person"},{"id":"dylan-scandinaro","name":"Dylan Scandinaro","type":"person"},{"id":"john-giannandrea","name":"John Giannandrea","type":"person"},{"id":"laurent-sifre","name":"Laurent Sifre","type":"person"},{"id":"openai-safety-security-committee","name":"OpenAI Safety and Security Committee","type":"organisation"},{"id":"john-diebold","name":"John Diebold","type":"person"},{"id":"chris-van-merwijk","name":"Chris van Merwijk","type":"person"},{"id":"baidu-20f-fy2023","name":"Baidu, Inc. annual report on Form 20-F for the fiscal year ended December 31, 2023","type":"document"},{"id":"ernie-5-report","name":"ERNIE 5.0 Technical Report","type":"document"},{"id":"iflytek","name":"iFlytek / 科大讯飞","type":"organisation"},{"id":"tc260-asgf-v1-0","type":"document","documentVersion":{"family":"tc260-ai-safety-governance-framework","label":"1.0","publisherId":"tc260","publishedAt":"2024-09-09","dateNote":"Released on 9 September 2024 at the National Cybersecurity Awareness Week forum, per TC260's news item; the cover gives only 2024年9月. No effective date is stated."}},{"id":"tc260-asgf-v2-0","name":"AI Safety Governance Framework 2.0 (人工智能安全治理框架2.0)","type":"document","documentVersion":{"family":"tc260-ai-safety-governance-framework","label":"2.0","publisherId":"tc260","publishedAt":"2025-09-15","dateNote":"Released on 15 September 2025 at the National Cybersecurity Awareness Week forum, per the CAC news item; the cover gives only 2025年9月. No effective date is stated."}},{"id":"tc260-asgf-v3-0","type":"document","documentVersion":{"family":"tc260-ai-safety-governance-framework","label":"3.0","publisherId":"tc260","publishedAt":"2026-09-14","dateNote":"Released on 14 September 2026 at the opening of National Cybersecurity Awareness Week, per TC260's news item; the cover gives 2026年9月. No effective date is stated."}},{"id":"uk-cohere-mou-2025","name":"Memorandum of Understanding between the UK and Cohere on AI opportunities","type":"document","documentVersion":{"family":"uk-cohere-mou","label":"GOV.UK text (published 16 June 2025)","publisherId":"dsit","publishedAt":"2025-06-16","dateNote":"Published on GOV.UK on 16 June 2025; no signing date is given."}},{"id":"us-eo-14299-2025","name":"Executive Order 14299: Deploying Advanced Nuclear Reactor Technologies for National Security","type":"document","documentVersion":{"family":"us-eo-14299","label":"As published in the Federal Register (29 May 2025)","publisherId":"us-white-house","publishedAt":"2025-05-29","dateNote":"Signed on 23 May 2025; published in the Federal Register of 29 May 2025."}},{"id":"constellation-energy","name":"Constellation Energy Corporation","type":"organisation"},{"id":"aib-alexander-rives","name":"Alexander Rives","type":"person"},{"id":"aib-doc-esm1b-2021","name":"Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences","type":"document"},{"id":"kp-gemini-family-multimodal-models-2023","name":"Gemini: A Family of Highly Capable Multimodal Models","type":"document"},{"id":"kp-opportunities-risks-foundation-models-2021","name":"On the Opportunities and Risks of Foundation Models","type":"document"},{"id":"kp-mmlu-pro-more-robust-challenging-2024","name":"MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark","type":"document"},{"id":"kp-robots-jobs-evidence-from-us-2019","name":"Robots and Jobs: Evidence from US Labor Markets","type":"document"},{"id":"kp-adversarial-winograd-schema-challenge-at-2019","name":"WinoGrande: An Adversarial Winograd Schema Challenge at Scale","type":"document"},{"id":"kp-mastering-game-go-deep-neural-2016","name":"Mastering the game of Go with deep neural networks and tree search","type":"document"},{"id":"kp-sequence-sequence-learning-neural-networks-2014","name":"Sequence to Sequence Learning with Neural Networks","type":"document"},{"id":"liu-qingfeng","name":"Liu Qingfeng","type":"person"},{"id":"ai-extinction-risk","name":"Risk of extinction or comparable catastrophe","type":"concept","proposition":{"code":"R07","criterion":"Advanced AI poses a risk of human extinction or of a comparable, irreversible global catastrophe, large enough to warrant priority among the problems facing humanity.","terms":["existential risk","risk of extinction","x-risk"]}},{"id":"alignment-is-hard","name":"Difficulty of alignment","type":"concept","proposition":{"code":"R06","criterion":"Making an advanced AI system reliably pursue the goals its developers intend (aligning it) is a hard, unsolved technical problem that will not be solved by default in the ordinary course of development.","terms":["alignment problem","control problem","value alignment"]}},{"id":"bender-koller-2020-climbing-nlu","name":"Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data","type":"document"},{"id":"hawking-russell-tegmark-wilczek-2014-independent","name":"Transcendence looks at the implications of artificial intelligence - but are we taking AI seriously enough?","type":"document"},{"id":"leike-2022-optimistic-alignment-approach","name":"Why I'm optimistic about our alignment approach","type":"document"},{"id":"llms-lack-agency","name":"Language models lack agency or understanding","type":"concept","proposition":{"code":"R14","criterion":"Large language models lack goals, agency or understanding of their own, being systems that model or reproduce text.","terms":["stochastic parrots","autocomplete","agency"]}},{"id":"nature-2023-editorial-stop-talking","name":"Stop talking about tomorrow's AI doomsday when AI poses risks today","type":"document"},{"id":"present-harms-priority","name":"Present harms deserve priority","type":"concept","proposition":{"code":"R12","criterion":"Documented present harms of AI systems (for example discrimination, exploitation of workers, surveillance, misinformation or concentration of power) deserve priority over hypothetical future catastrophic risks, or attention to the latter diverts attention from the former.","terms":["present harms","current harms","existing harms"]}}],"claims":[{"id":"babuschkin-openai-prior","subject":"igor-babuschkin","predicate":"held_role","object":{"kind":"entity","entityId":"openai"},"topic":"people","statement":"Igor Babuschkin’s career account describes work on reasoning and pretraining at OpenAI.","time":{"kind":"unknown","note":"The source establishes a prior role but does not supply employment dates. No start or end is inferred."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"No start/end dates, duration, or precise job title are supplied. No departure event is inferred.","evidenceIds":["evidence-babuschkin-openai-prior"],"attributes":{"role":"Researcher (specific rank unstated)"}},{"id":"rsp-v2-1-rd-thresholds","subject":"rsp-v2-1","predicate":"changes_from","object":{"kind":"entity","entityId":"rsp-v2-0"},"topic":"governance","statement":"RSP v2.1 separates AI R&D thresholds for automating entry-level research and dramatically accelerating effective scaling.","time":{"kind":"event","date":"2025-03-31","basis":"effective","note":"The stated effective date is retained; a separate publication day is not inferred."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"This is a scoped change in the policy text. It does not establish implementation, compliance, risk reduction or a comparative safety ranking.","evidenceIds":["evidence-rsp-v2-1-rd-thresholds"],"attributes":{"changeBasis":"changelog"}},{"id":"sutskever-ssi-ceo-2025","subject":"ilya-sutskever","predicate":"held_role","object":{"kind":"entity","entityId":"ssi"},"topic":"people","statement":"SSI’s July 2025 signed update identifies Ilya Sutskever as CEO.","time":{"kind":"snapshot","date":"2025-07-03","note":"Role snapshot at the dated source; no start, end or continuing appointment inferred."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Historical corporate responsibility snapshot; no current role or earlier CEO start date inferred.","evidenceIds":["evidence-sutskever-ssi-ceo-2025"],"attributes":{"role":"CEO"}},{"id":"cn-r1-rl","subject":"deepseek-r1-report","predicate":"discusses","object":{"kind":"entity","entityId":"reinforcement-learning"},"topic":"ideas","statement":"The R1 report distinguishes R1-Zero’s RL without preliminary SFT from R1’s cold-start training.","time":{"kind":"event","date":"2025-01-22","basis":"publication","note":"Date of this source expression or announcement, not a separately verified implementation date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Authors’ method account; no independent reproduction or earliest-invention claim.","evidenceIds":["evidence-cn-r1-rl"]},{"id":"prov-rsp-v3-4-competitor-lead-delay","subject":"rsp-v3-4","predicate":"states_provision","object":{"kind":"literal","value":"Delay development and deployment when Anthropic is in the lead"},"topic":"governance","statement":"RSP v3.4 states that, in the scenario where Anthropic is in the lead, it will require a strong argument that catastrophic risk is contained, along the lines of its industry-wide recommendations, and will delay AI development and deployment as needed to achieve this until and unless it no longer believes it has a significant lead.","time":{"kind":"event","date":"2026-07-08","basis":"effective","note":"Provision as stated in this edition, dated by the edition’s effective date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the edition states, quoted from the cited expression. It is not evidence of implementation, compliance or effectiveness.","evidenceIds":["evidence-prov-rsp-v3-4-competitor-lead-delay"],"attributes":{"provision":{"kind":"required_action","key":"rsp-competitor-lead-delay"}}},{"id":"cs-anthropic-ltbt-removes-directors","subject":"anthropic-ltbt","predicate":"holds_control_right","object":{"kind":"entity","entityId":"anthropic"},"topic":"governance","statement":"The restated certificate filed on 3 July 2025 provides that any director elected under Article VIII(B) may be removed at any time without cause by the holders of the class or series of stock entitled to elect that director, so the holders of Class T Common Stock, which Anthropic states is held by the Long-Term Benefit Trust, may remove the Class T Directors.","time":{"kind":"snapshot","date":"2025-07-03","note":"Filing date of the latest inspected certificate; the same provision is in the certificate filed on 31 March 2023."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the filed text says, applied to the Class T Directors it defines; the Trust’s holding rests on Anthropic’s statements. Removal for cause “as provided by law” is a separate route that is not interpreted here.","evidenceIds":["evidence-cs-anthropic-ltbt-removes-directors-1","evidence-cs-anthropic-ltbt-removes-directors-2"],"attributes":{"control":{"right":"removes_directors","instrument":"Class T Common Stock (certificate of incorporation, Article VIII(B)7)","stage":"effective"}}},{"id":"prov-wh-2023-07-public-reporting","subject":"wh-voluntary-ai-commitments-2023-07","predicate":"states_provision","object":{"kind":"literal","value":"Commitment 6: public reports for significant model releases in scope"},"topic":"governance","statement":"Commitment 6 of the July 2023 text commits companies to publish reports for all new significant public model releases within scope, covering safety evaluations conducted, significant limitations in performance, effects on societal risks and the results of adversarial testing.","time":{"kind":"event","date":"2023-07-21","basis":"publication","note":"The edition’s publication date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the text says, not evidence that reports were published.","evidenceIds":["evidence-prov-wh-2023-07-public-reporting-1"],"attributes":{"provision":{"kind":"required_action","key":"public-reporting"}}},{"id":"cm-seoul-xai","subject":"xai","predicate":"signed_commitment","object":{"kind":"entity","entityId":"frontier-ai-safety-commitments-2024"},"topic":"governance","statement":"xAI is listed among the sixteen organisations that, according to the GOV.UK text of 21 May 2024, the UK and Republic of Korea governments announced had agreed to the Frontier AI Safety Commitments.","time":{"kind":"event","date":"2024-05-21","basis":"publication","note":"First publication of the GOV.UK text, on the opening day of the AI Seoul Summit; the Internet Archive capture of 2024-05-21 has the sixteen-name list."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Listing is the convener’s statement. It is not evidence that the organisation complied with, implemented or still maintains the commitments, and the legal effect of being listed is not recorded.","evidenceIds":["evidence-cm-seoul-xai-1"],"attributes":{"commitment":{"stage":"listed"}}},{"id":"ts-b6-vinge-1983-context","subject":"vinge-1983-first-word","predicate":"bears_on_proposition","object":{"kind":"entity","entityId":"control-or-purpose-specification"},"topic":"ideas","statement":"Vinge’s 1983 column expects human leadership to end, while also considering partnerships between people and computers.","time":{"kind":"event","date":"1983-01","basis":"publication","note":"Omni, January 1983, p. 10, according to Vinge’s 2005 bibliography; the scan shows the byline, column title and page but no masthead."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A forecast of displacement, not of a failure of control.","evidenceIds":["evidence-ts-b6-vinge-1983-context"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Human leadership is predicted to end, without a claim that people could not direct, correct or stop the machines."}}},{"id":"yk-y08-sl4-2000-07-09-yudkowsky-0007-0017-b","subject":"sl4-2000-07-09-yudkowsky-0007-0017","predicate":"bears_on_proposition","object":{"kind":"entity","entityId":"warning-signs-in-time"},"topic":"ideas","statement":"The SL4 message “Re: [SL4] Programmed morality” (2000-07-09) bears on Y08. Replying to Dale Johnstone on whether a developing AI should have Web access, he says current nervousness about rogue AI reflects lack of experience, and that working with AI before it becomes a threat will give an excellent model of its motivations.","time":{"kind":"event","date":"2000-07-09","basis":"publication","note":"Archive author index lists the message as \"Sun Jul 09 2000 - 13:53:37 MDT\" under \"Eliezer S. Yudkowsky\"."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Dangerous failure modes will show up first in weaker systems, where they can be observed and fixed. Earlier view (July 2000). Later rejection located in this archive: his SL4 message of 2006-06-06, which says observed good behaviour in a prehuman AI does not license the conclusion that it will stay good; the same message (paragraph beginning 'Now in practice') adds that to interpret an AGI's behaviour 'you may need to solve most of the problem in order to know what you're seeing'. The 2006 message rejects the proposition but does not refer back to his 2000 statement; the reviewer should confirm whether this meets the earlier-self standard or downgrade to related. Proposed as an earlier view; recorded as related because no later text located so far explicitly rejects this earlier statement. An earlier view; rejected explicitly in 2006 (his SL4 message of 2006-06-06), which does not refer back to this message. Cross-family check: confidence in a future model of AI motivations is not the claim that dangerous failures first appear in weaker systems; the 2006 text rejects extrapolation from behaviour, a different proposition.","evidenceIds":["evidence-yk-t08-s1"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Confidence in a future understanding of AI motivations; it does not claim that dangerous failure modes first appear in weaker systems."}}},{"id":"yk-y08-sl4-2006-06-06-yudkowsky-0606-15174-c","subject":"sl4-2006-06-06-yudkowsky-0606-15174","predicate":"contests_proposition","object":{"kind":"entity","entityId":"warning-signs-in-time"},"topic":"ideas","statement":"The SL4 message “Re: [agi] Two draft papers: AI and existential risk; heuristics and biases” (2006-06-06) contests Y08. Replying to Ben Goertzel's call for 'AGI experimentation' as an answer to Friendly AI concerns, he says there is a major context change between the prehuman and posthuman stages, so observed good behaviour does not show that the AI will stay good.","time":{"kind":"event","date":"2006-06-06","basis":"publication","note":"Archive author index lists the message as \"Tue Jun 06 2006 - 23:19:49 MDT\" under \"Eliezer S. Yudkowsky\"."},"evidenceIds":["evidence-yk-t08-s5"]},{"id":"yk-y13-sl4-2002-07-02-yudkowsky-0207-4667-b","subject":"sl4-2002-07-02-yudkowsky-0207-4667","predicate":"bears_on_proposition","object":{"kind":"entity","entityId":"agi-decades-away"},"topic":"ideas","statement":"The SL4 message “Re: Seed AI (was: How hard a Singularity?)” (2002-07-02) bears on Y13. Replying to Samantha Atkins, who questions his earlier warning against waiting until an AI can read humanity's moral literature, he says Friendliness work started at that stage would come too late and must be present from the beginning.","time":{"kind":"event","date":"2002-07-02","basis":"publication","note":"Archive author index lists the message as \"Tue Jul 02 2002 - 23:01:42 MDT\" under \"Eliezer S. Yudkowsky\"."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Long timelines mean there is no need to act now. About when Friendliness work must start relative to the AI's development, not about calendar timelines. The sentence begins \"An AI advanced enough to understand world literature and philosophy regarding morality is an AI more than advanced enough that\"; the point is developmental stage, not calendar time.","evidenceIds":["evidence-yk-t13-s3"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"About the stage of an AI's development at which Friendliness work must start, not about calendar timelines."}}},{"id":"ec-g09-erdil-besiroglu-2023-explosive-growth-review-a","subject":"erdil-besiroglu-2023-explosive-growth-review","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"explosive-growth-estimates"},"topic":"ideas","statement":"Erdil and Besiroglu, “Explosive growth from AI automation: A review of the arguments” (2023-10-01) articulates G09. The authors' unconditional estimate. The same paragraph says high confidence \"seems unwarranted\".","time":{"kind":"event","date":"2023-10-01","basis":"publication","note":"Date of v2, the checked version. arXiv submission history (2026 capture of the abstract page): v1 2023-09-20, v2 2023-10-01, v3 2024-07-15."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Definition corrected after the cross-family check.","evidenceIds":["evidence-ec-g09-f"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","claimed":{"date":"2023-09-20","note":"Date of v1, which was not retrieved."},"attested":{"date":"2023-10-01","note":"Version 2 (2023-10-01) is the earliest retrieved version; v1 has no capture. The v2 PDF itself was captured in 2025."},"tier":{"value":1,"basis":"Tier 1 (re-tiered on 2026-10-25). A scholarly review with an economist author: Tamay Besiroglu held an MPhil in Economics from the University of Cambridge before the text, as his own site says in a capture of 2021-09-15 and his MIT FutureTech profile states (degree of 2020); MIT FutureTech is his affiliation on the v2 title page. This reverses the tier-2 borderline settled on 2026-10-11, whose premise, that neither author was shown to be an economist, did not hold."},"conditions":"Explosive growth defined in the paper as annual real GWP exceeding 130% of its maximum value over all previous years (about ten times growth rates typical of today's frontier economies).","estimates":[{"holder":"Ege Erdil and Tamay Besiroglu","value":"about even odds of widespread automation and subsequent explosive growth","horizon":"by the end of this century (from October 2023)","definition":"annual real gross world product exceeding 130% of its maximum over all previous years","conditional":false,"conditions":"The same paragraph says high confidence \"seems unwarranted\"."}]}}},{"id":"ec-m01-leontief-duchin-1984-impacts-of-automation-b","subject":"leontief-duchin-1984-impacts-of-automation","predicate":"bears_on_proposition","object":{"kind":"entity","entityId":"automation-as-ordinary-technical-change"},"topic":"ideas","statement":"Leontief and Duchin, “The Impacts of Automation on Employment, 1963–2000. Final Report” (1984-04) bears on M01. Speaker: Wassily Leontief and Faye Duchin (economists). Puts the 1984 projections in a long-run perspective.","time":{"kind":"event","date":"1984-04","basis":"publication","note":"ERIC résumé: \"Apr 84\". Published in book form as The Future Impact of Automation on Workers (Oxford University Press, 1986; not retrieved). Chapters 4 and 5 were drafted by other members of the Institute (Howell; McDonough), per the report; the record attributes the report to Leontief and Duchin as its authors. Checked in a current copy; an archive capture was not available, so earlier bytes are not authenticated."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Matches the cached .txt.","evidenceIds":["evidence-ec-m01-11"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"A historical analogy that treats the 'electronic revolution' like mechanical power. It is not an analytical treatment of computers as ordinary technical change, though the report's method (technical coefficients in an input-output model) does treat them that way.","tier":{"value":1,"basis":"Report by economists (Leontief and Duchin, NYU Institute for Economic Analysis)."},"conditions":"Preceded by: the industrial revolution of mechanical power continued 'over a period of some two hundred years'."}}},{"id":"ec-m09-agrawal-gans-goldfarb-2018-economic-policy-for-ai-b","subject":"agrawal-gans-goldfarb-2018-economic-policy-for-ai","predicate":"bears_on_proposition","object":{"kind":"entity","entityId":"recent-economics-engagement-with-ai"},"topic":"ideas","statement":"Agrawal, Gans, Goldfarb, “Economic Policy for Artificial Intelligence” (2018-06) bears on M09. Summarises economists' long-run optimism on jobs while stressing short-run adjustment costs.","time":{"kind":"event","date":"2018-06","basis":"publication","note":"Cover \"June 2018\". Checked in a current copy; an archive capture was not available, so earlier bytes are not authenticated."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"NBER Working Paper 24690, June 2018. The authors' characterisation of a consensus.","evidenceIds":["evidence-ec-m09-23"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Characterises economists' views on AI and jobs; says nothing about when the field took up AI.","tier":{"value":1,"basis":"NBER working paper (core venue)."}}}},{"id":"lg-oai-ssc-formed-2024","subject":"openai-safety-security-committee","predicate":"unit_of","object":{"kind":"entity","entityId":"openai-inc"},"topic":"governance","statement":"OpenAI announced on 28 May 2024 that its board had formed a Safety and Security Committee to make recommendations to the full board on critical safety and security decisions.","time":{"kind":"snapshot","date":"2024-05-28","note":"Announcement date of the committee’s formation."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The announcement does not name the legal entity; the record assigns the board to OpenAI, Inc., the nonprofit that OpenAI’s 2019 note names as OpenAI Inc and that, according to OpenAI’s structure page, has always governed and controlled the for-profit (context evidence). A recommending committee as announced, not evidence of how it has operated.","evidenceIds":["lg-ev-oai-ssc-formed-2024","lg-ev-oai-ssc-formed-2024-inc","lg-ev-oai-ssc-formed-2024-governed"]},{"id":"lg-oai-scandinaro-preparedness-2026","subject":"dylan-scandinaro","predicate":"held_role","object":{"kind":"entity","entityId":"openai"},"topic":"people","statement":"Sam Altman and Dylan Scandinaro each posted on X on 3 February 2026 that Scandinaro was joining OpenAI as its Head of Preparedness.","time":{"kind":"snapshot","date":"2026-02-03","note":"Date of both posts; not a start date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An announced appointment; neither post gives a start date. Scandinaro’s post refers to his time at Anthropic without naming a role there.","evidenceIds":["lg-ev-oai-scandinaro-preparedness-2026-1","lg-ev-oai-scandinaro-preparedness-2026-2"],"attributes":{"activityStage":"announced","role":"Head of Preparedness"}},{"id":"lg-ant-mann-openai-prior","subject":"ben-mann","predicate":"held_role","object":{"kind":"entity","entityId":"openai"},"topic":"people","statement":"An Anthropic founder slide of 2024 describes Ben Mann as a former GPT-3 and API engineer at OpenAI.","time":{"kind":"unknown","note":"The slide does not date the role."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Anthropic's description of a prior role; no dates, title history or departure date are established.","evidenceIds":["lg-ev-ant-mann-openai-prior"],"attributes":{"role":"GPT-3 and API Engineer"}},{"id":"lg-cn-zhang-peng-zhipu-cofounder","subject":"zhang-peng","predicate":"co_founded","object":{"kind":"entity","entityId":"zhipu"},"topic":"people","statement":"Zhipu’s Hong Kong prospectus of 30 December 2025 describes Zhang Peng as a co-founder of the company.","time":{"kind":"unknown","note":"The prospectus gives the co-founder designation without a personal founding date; the company’s establishment date is recorded separately."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A retrospective co-founder designation in the company’s own filing; it does not establish the person’s role at founding or a continuing role. The prospectus names six co-founders; the dataset does not establish that the list is exhaustive.","evidenceIds":["lg-ev-cn-zhang-peng-zhipu-cofounder"]},{"id":"cmp-preparedness-beta-2023-d3-safeguards","subject":"preparedness-beta-2023","predicate":"states_provision","object":{"kind":"literal","value":"Safeguards required"},"topic":"governance","statement":"Preparedness · Beta states a provision on the comparison question “Safeguards required”.","time":{"kind":"event","date":"2023-12-18","basis":"publication","note":"Comparison cell for this edition, dated by the edition’s publication date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The other two safety baselines (restricting deployment and development, PDF p. 21) require mitigations that bring post-mitigation risk to set levels; they are recorded under the pause facet. What the edition states, quoted from the cited expression. It is not evidence of implementation, compliance or effectiveness, and a shared question does not make two editions’ provisions equivalent.","evidenceIds":["ev-cmp-preparedness-beta-2023-d3-safeguards-1"],"attributes":{"provision":{"kind":"required_action","key":"comparison-d3-safeguards","facet":{"id":"D3.safeguards","status":"stated"}}}},{"id":"ap-john-diebold-diebold-1955-automation-condensation","subject":"john-diebold","predicate":"authored","object":{"kind":"entity","entityId":"diebold-1955-automation-condensation"},"topic":"ideas","statement":"“A Condensation of the Book Automation”, printed in the 1955 hearings Automation and Technological Change (1955-10-14), condenses John Diebold’s book Automation; the condensation was prepared by the Controllership Foundation.","time":{"kind":"event","date":"1955-10-14","basis":"publication","note":"Date the source record gives for this text; authorship is credited in the inspected expression."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Diebold is the author of the underlying book; the condensation itself was prepared by the Controllership Foundation and inserted into the record with his statement, as the existing records on it say. The book’s wording was not compared. It says nothing about priority for the ideas discussed.","evidenceIds":["evidence-ap-john-diebold-diebold-1955-automation-condensation"]},{"id":"ap-chris-van-merwijk-rlo-2019-deceptive-alignment","subject":"chris-van-merwijk","predicate":"authored","object":{"kind":"entity","entityId":"rlo-2019-deceptive-alignment"},"topic":"ideas","statement":"“Deceptive Alignment (Risks from Learned Optimization, part 4)” (2019-06-05) credits Chris van Merwijk as one of its 5 authors.","time":{"kind":"event","date":"2019-06-05","basis":"publication","note":"Date the source record gives for this text; authorship is credited in the inspected expression."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Credited co-authorship; it does not allocate parts of the text between authors or imply agreement with every passage.","evidenceIds":["evidence-ap-chris-van-merwijk-rlo-2019-deceptive-alignment"]},{"id":"ap-eliezer-yudkowsky-extropians-1998-09-07-yudkowsky-2209","subject":"eliezer-yudkowsky","predicate":"authored","object":{"kind":"entity","entityId":"extropians-1998-09-07-yudkowsky-2209"},"topic":"ideas","statement":"The mailing-list message “Human AI to superhuman (Re: Max More)” (1998-09-07) is sent under the name Eliezer S. Yudkowsky.","time":{"kind":"event","date":"1998-09-07","basis":"publication","note":"Date the source record gives for this text; authorship is credited in the inspected expression."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A sender name on a mailing-list message as archived; it does not establish identity beyond the credited name, and quoted text by others inside the message is not the sender’s.","evidenceIds":["evidence-ap-eliezer-yudkowsky-extropians-1998-09-07-yudkowsky-2209"]},{"id":"ap-eliezer-yudkowsky-sl4-2002-06-23-yudkowsky-0206-4219","subject":"eliezer-yudkowsky","predicate":"authored","object":{"kind":"entity","entityId":"sl4-2002-06-23-yudkowsky-0206-4219"},"topic":"ideas","statement":"The SL4 message “Re: Seed AI (was: How hard a Singularity?)” of 2002-06-23, archive message 4219, is sent under the name Eliezer S. Yudkowsky.","time":{"kind":"event","date":"2002-06-23","basis":"publication","note":"Date the source record gives for this text; authorship is credited in the inspected expression."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A sender name on a mailing-list message as archived; it does not establish identity beyond the credited name, and quoted text by others inside the message is not the sender’s.","evidenceIds":["evidence-ap-eliezer-yudkowsky-sl4-2002-06-23-yudkowsky-0206-4219"]},{"id":"cmp-rsp-v2-1-d8-process","subject":"rsp-v2-1","predicate":"states_provision","object":{"kind":"literal","value":"Revision process and approval"},"topic":"governance","statement":"RSP · v2.1 states a provision on the comparison question “Revision process and approval”.","time":{"kind":"event","date":"2025-03-31","basis":"effective","note":"Comparison cell for this edition, dated by the edition’s effective date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the edition states, paraphrased from the cited expression with its page. It is not evidence of implementation, compliance or effectiveness, and a shared question does not make two editions’ provisions equivalent.","evidenceIds":["ev-cmp-rsp-v2-1-d8-process-1"],"attributes":{"provision":{"kind":"revision","key":"comparison-d8-process","facet":{"id":"D8.process","status":"stated"}}}},{"id":"c2-baidu-20f-fy2023-genai-filing","subject":"baidu","predicate":"published","object":{"kind":"entity","entityId":"baidu-20f-fy2023"},"topic":"governance","statement":"Baidu's annual report on Form 20-F, filed on 15 March 2024, states that Baidu completed the security assessment and algorithm filing required under the Generative AI Services Measures and other applicable rules before launching ERNIE Bot to the public.","time":{"kind":"event","date":"2024-03-15","basis":"publication","note":"Filing date of the annual report; the filing does not date the assessment or the algorithm filing."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Baidu's own statement in a securities filing; the record does not establish the content or outcome of any regulator's review, or that this statement was independently verified. The filing's own term 'Generative AI Services Measures' is kept as written.","evidenceIds":["c2-ev-baidu-20f-fy2023-genai-filing"]},{"id":"c2-governance-tc260-asgf-v2-loss-of-control-testing","subject":"tc260-asgf-v2-0","predicate":"states_provision","object":{"kind":"literal","value":"Developers to test models regularly for potential loss-of-control risk"},"topic":"governance","statement":"Under fostering consensus on loss-of-control risk, the AI Safety Governance Framework 2.0 states that developers regularly test whether a model may bring potential technical loss-of-control risk.","time":{"kind":"event","date":"2025-09-15","basis":"publication","note":"Release date of the cited edition; the provision's own effect or application date is not stated."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What version 2.0 says. The sentence has no modal verb; it is recorded as a stated measure, not as an obligation on any named developer, and no test protocol or frequency is given.","evidenceIds":["c2-ev-governance-tc260-asgf-v2-loss-of-control-testing-1"],"attributes":{"provision":{"kind":"required_action","key":"loss-of-control-testing"}}},{"id":"fl-eu-aia-art52-5-reassessment","subject":"eu-ai-act-2024","predicate":"states_provision","object":{"kind":"literal","value":"Reassessment of a designation on reasoned request, at the earliest six months after the decision"},"topic":"governance","statement":"Article 52(5) allows a provider whose model was designated under Article 52(4) to request reassessment on objective, detailed and new reasons, at the earliest six months after the designation decision, and again six months after any decision to maintain it.","time":{"kind":"event","date":"2024-07-12","basis":"publication","note":"Provision as published in the Official Journal on 2024-07-12 (the Regulation is dated 13 June 2024). Chapter V (Articles 51 to 56) applies from 2 August 2025 (Article 113, third paragraph, point (b); recorded separately)."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"States what the Regulation requires of the categories it defines. Whether any named developer's model falls within a category, and whether any developer complies, is legal application and is not recorded. A statutory duty is not evidence of implementation or enforcement. Regulation (EU) 2026/1744, inspected 2026-10-27, does not list this article among its amendments (its Article 1 amends Articles 1-6, 10, 11, 17, 25, 27-30, 40, 42, 43, 50, 56(6), 57, 58, 60, 63, 64, 69, 70, 72, 75-77, 95-97, 99, 111, 113 and Annexes I and VIII).","evidenceIds":["evidence-fl-eu-aia-art52-5-reassessment-1","evidence-fl-eu-aia-art52-5-reassessment-2"],"attributes":{"provision":{"kind":"exception","key":"aia-art52-5"}}},{"id":"fl-uk-cohere-mou-not-binding","subject":"uk-cohere-mou-2025","predicate":"states_provision","object":{"kind":"literal","value":"The memorandum is voluntary, not legally binding and without prejudice to any binding agreements."},"topic":"governance","statement":"The UK-Cohere Memorandum of Understanding states that it is voluntary, not legally binding and without prejudice to any binding agreements.","time":{"kind":"event","date":"2025-06-16","basis":"publication","note":"Date of publication, as the source gives it."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The text's own statement of status.","evidenceIds":["evidence-fl-uk-cohere-mou-not-binding-1"],"attributes":{"provision":{"kind":"exception","key":"uk-cohere-mou-not-binding"}}},{"id":"cpo-pw-us-eo14299-reactors-for-ai","subject":"us-eo-14299-2025","predicate":"states_provision","object":{"kind":"literal","value":"Sec. 4(c): privately funded advanced reactors at DOE sites to power AI infrastructure; first reactor operating within 30 months"},"topic":"governance","statement":"Section 4(c) of Executive Order 14299 directed the Secretary of Energy to use available authorities to site and authorise privately funded advanced reactors at DOE sites for purposes including powering AI infrastructure, with a goal of operating a reactor at the first site within 30 months of the order.","time":{"kind":"event","date":"2025-05-29","basis":"publication","note":"Provision as the text states it, dated by its date of publication, 2025-05-29. Signed on 23 May 2025; published in the Federal Register of 29 May 2025."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A goal stated in an order (about November 2027 by calculation from the signature date, not stated as a date in the text); no reactor operation is recorded.","evidenceIds":["evidence-cpo-pw-us-eo14299-reactors-for-ai-1","evidence-cpo-pw-us-eo14299-reactors-for-ai-2"],"attributes":{"provision":{"kind":"required_action","key":"us-eo14299-sec4c"}}},{"id":"cpo-pw-google-constellation-890mw","subject":"google","predicate":"power_agreement","object":{"kind":"entity","entityId":"constellation-energy"},"topic":"compute","statement":"Google and Constellation announced on 6 October 2026 a 20-year power purchase agreement to bring 890 MW of new nuclear capacity onto the PJM grid through uprates at 11 Constellation-owned units in Illinois, Pennsylvania and New Jersey, with the first uprate expected by 2028.","time":{"kind":"event","date":"2026-10-06","basis":"announcement","note":"Date of the announcement, as the source gives it."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Announced; 890 MW is incremental uprate capacity to be built, not power delivered. The release ties the agreement to Google's 'growth' in PJM, not to a named data centre. The concurrent 15-year 2,700 MW supply agreement is a separate record (recorded separately).","evidenceIds":["evidence-cpo-pw-google-constellation-890mw-1","evidence-cpo-pw-google-constellation-890mw-2","evidence-cpo-pw-google-constellation-890mw-3","evidence-cpo-pw-google-constellation-890mw-4"],"attributes":{"transactionStage":"contracted","quantity":{"value":890,"unit":"megawatts (MW) of new nuclear capacity","bound":"exact","measures":"generation_capacity"},"statedDates":[{"kind":"delivery_target","date":"2028","wording":"first uprate","statedBy":"constellation-energy"}],"power":{"technology":"nuclear","mechanism":"ppa"},"term":"20 years"}},{"id":"pp-er-giannandrea-apple-svp-2025","subject":"john-giannandrea","predicate":"held_role","object":{"kind":"entity","entityId":"apple"},"topic":"people","statement":"Apple’s announcement of 1 December 2025 describes John Giannandrea as its senior vice president for Machine Learning and AI Strategy.","time":{"kind":"snapshot","date":"2025-12-01","note":"Date of the announcement that he was stepping down; the role's start is not given in the passage."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A role snapshot from the departure announcement, which is the dataset's separate record; it does not date the end of the role.","evidenceIds":["pp-ev-er-giannandrea-apple-svp-2025-1"],"attributes":{"role":"Senior Vice President for Machine Learning and AI Strategy"}},{"id":"aib-ppl-rives-esm1b","subject":"aib-alexander-rives","predicate":"authored","object":{"kind":"entity","entityId":"aib-doc-esm1b-2021"},"topic":"ideas","statement":"Alexander Rives is credited as the lead-listed author of ‘Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences’ (PNAS, 2021).","time":{"kind":"event","date":"2021-04-05","basis":"publication","note":"Date in the PubMed Central header (Proc Natl Acad Sci U S A. 2021 Apr 5;118(15)); the kind of this date is not stated there and the paper is dated by its own record."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Co-authorship as credited; not sole credit, and not a statement about any individual's contribution beyond what is quoted. The paper lists affiliations with Facebook AI Research and New York University; these are not recorded as employment.","evidenceIds":["evidence-aib-ppl-rives-esm1b-1","evidence-aib-ppl-rives-esm1b-2"]},{"id":"kp-states-mmlu-pro-more-robust-challenging-2024-1","subject":"kp-mmlu-pro-more-robust-challenging-2024","predicate":"states_contribution","object":{"kind":"literal","value":"The paper introduces MMLU-Pro, a dataset that extends the MMLU benchmark with more reasoning-focused questions and expands the answer choices from four to ten."},"topic":"ideas","statement":"The paper introduces MMLU-Pro, a dataset that extends the MMLU benchmark with more reasoning-focused questions and expands the answer choices from four to ten.","time":{"kind":"event","date":"2024-06-03","basis":"publication","note":"Date of the arXiv version quoted, submitted 3 June 2024 (UTC); later versions may word it differently."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The paper's own description of its contribution; not an assessment of it.","evidenceIds":["evidence-kp-states-mmlu-pro-more-robust-challenging-2024-1-1"]},{"id":"kp-states-adversarial-winograd-schema-challenge-at-2019-1","subject":"kp-adversarial-winograd-schema-challenge-at-2019","predicate":"states_contribution","object":{"kind":"literal","value":"The paper introduces WinoGrande, a collection of Winograd Schema Challenge problems that it states are adversarially constructed to be robust against spurious statistical biases."},"topic":"ideas","statement":"The paper introduces WinoGrande, a collection of Winograd Schema Challenge problems that it states are adversarially constructed to be robust against spurious statistical biases.","time":{"kind":"event","date":"2019-07-24","basis":"publication","note":"Date of the arXiv version quoted, submitted 24 July 2019 (UTC); later versions may word it differently."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The paper's own description of its contribution; not an assessment of it.","evidenceIds":["evidence-kp-states-adversarial-winograd-schema-challenge-at-2019-1-1"]},{"id":"kp-cites-iem-2013-ai-positive-negative-factor-global-risk","subject":"iem-2013","predicate":"cites","object":{"kind":"entity","entityId":"ai-positive-negative-factor-global-risk"},"topic":"ideas","statement":"Intelligence Explosion Microeconomics cites Artificial Intelligence as a Positive and Negative Factor in Global Risk.","time":{"kind":"event","date":"2013","basis":"publication","note":"Date of the citing source as recorded in the dataset; unknown where the dataset has none."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An explicit reference only; neither agreement nor influence. The citing text does not identify which version of the work was read.","evidenceIds":["evidence-kp-cites-iem-2013-ai-positive-negative-factor-global-risk-1"]},{"id":"kp-cites-ernie-5-report-deepseek-r1-incentivizing-reasoning-capability-2025","subject":"ernie-5-report","predicate":"cites","object":{"kind":"entity","entityId":"deepseek-r1-report"},"topic":"ideas","statement":"ERNIE 5.0 Technical Report cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.","time":{"kind":"event","date":"2026-02-04","basis":"publication","note":"Date of the citing source as recorded in the dataset; unknown where the dataset has none."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An explicit reference only; neither agreement nor influence. The citing text does not identify which version of the work was read.","evidenceIds":["evidence-kp-cites-ernie-5-report-deepseek-r1-incentivizing-reasoning-capability-2025-1"]},{"id":"kp-cites-acemoglu-restrepo-2019-wrong-kind-of-ai-robots-jobs-evidence-from-us-2019","subject":"acemoglu-restrepo-2019-wrong-kind-of-ai","predicate":"cites","object":{"kind":"entity","entityId":"kp-robots-jobs-evidence-from-us-2019"},"topic":"ideas","statement":"The Wrong Kind of AI? Artificial Intelligence and the Future of Labor Demand cites Robots and Jobs: Evidence from US Labor Markets.","time":{"kind":"event","date":"2019-03","basis":"publication","note":"Date of the citing source as recorded in the dataset; unknown where the dataset has none."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An explicit reference only; neither agreement nor influence. The citing text does not identify which version of the work was read.","evidenceIds":["evidence-kp-cites-acemoglu-restrepo-2019-wrong-kind-of-ai-robots-jobs-evidence-from-us-2019-1"]},{"id":"kp-cites-source-gemini-report-v1-sequence-sequence-learning-neural-networks-2014","subject":"kp-gemini-family-multimodal-models-2023","predicate":"cites","object":{"kind":"entity","entityId":"kp-sequence-sequence-learning-neural-networks-2014"},"topic":"ideas","statement":"Gemini: A Family of Highly Capable Multimodal Models cites Sequence to Sequence Learning with Neural Networks.","time":{"kind":"event","date":"2023-12-19","basis":"publication","note":"Date of the citing source as recorded in the dataset; unknown where the dataset has none."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An explicit reference only; neither agreement nor influence. The citing text does not identify which version of the work was read.","evidenceIds":["evidence-kp-cites-source-gemini-report-v1-sequence-sequence-learning-neural-networks-2014-1"]},{"id":"kp-authored-laurent-sifre-mastering-game-go-deep-neural-2016","subject":"laurent-sifre","predicate":"authored","object":{"kind":"entity","entityId":"kp-mastering-game-go-deep-neural-2016"},"topic":"ideas","statement":"Laurent Sifre is a credited author of Mastering the game of Go with deep neural networks and tree search.","time":{"kind":"event","date":"2016","basis":"publication","note":"Date of the work’s earliest recorded version (2016), at the precision its source gives."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Authorship as credited in the byline of the text read; it implies neither sole authorship nor a role, and no affiliation is recorded.","evidenceIds":["evidence-kp-authored-laurent-sifre-mastering-game-go-deep-neural-2016-1"]},{"id":"kp-authored-dario-amodei-amplification-paper","subject":"dario-amodei","predicate":"authored","object":{"kind":"entity","entityId":"amplification-paper"},"topic":"ideas","statement":"Dario Amodei is a credited author of Supervising strong learners by amplifying weak experts.","time":{"kind":"event","date":"2018-10-19","basis":"publication","note":"Date of the work’s earliest recorded version, arXiv version 1, submitted 19 October 2018 (UTC)."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Authorship as credited in the byline of the text read; it implies neither sole authorship nor a role, and no affiliation is recorded.","evidenceIds":["evidence-kp-authored-dario-amodei-amplification-paper-1"]},{"id":"th-term-fm-bommasani-2021-naming","subject":"kp-opportunities-risks-foundation-models-2021","predicate":"states_provision","object":{"kind":"literal","value":"The report's authors chose the new term \"foundation models\" after surveying existing terms such as \"pretrained model\" and \"self-supervised model\"."},"topic":"ideas","statement":"Bommasani et al. (2021) state that, having surveyed existing terms such as \"pretrained model\" and \"self-supervised model\", they chose to coin the new term \"foundation models\".","time":{"kind":"event","date":"2021-08-16","basis":"publication","note":"Dated by the text that states it, 16 August 2021. The statement reports 16 August 2021: Date of submission (an arXiv version or a journal’s received date, as the source states it); not a later revision or the venue’s publication date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The text's own priority statement (\"coin the new term\"), recorded as its statement; Selekoa does not adopt it. The section heading is \"Naming\" (1.1.1).","evidenceIds":["evidence-th-term-fm-bommasani-2021-naming-1","evidence-th-term-fm-bommasani-2021-naming-2"],"attributes":{"provision":{"kind":"statement","key":"term-fm-bommasani-2021-naming"}}},{"id":"c3-iflytek-liu-qingfeng-president-2008","subject":"liu-qingfeng","predicate":"held_role","object":{"kind":"entity","entityId":"iflytek"},"topic":"people","statement":"iFlytek’s 2008 IPO prospectus lists Liu Qingfeng as a director and president (总裁) of the company.","time":{"kind":"snapshot","date":"2008-04-24","note":"Roles shown in the prospectus published on 24 April 2008; not extended to any later date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"总裁 is rendered “president”.","evidenceIds":["c3-ev-iflytek-liu-qingfeng-president-2008"],"attributes":{"role":"董事、总裁 (director and president)"}},{"id":"rd-r06-leike-2022-optimistic-alignment-approach","subject":"leike-2022-optimistic-alignment-approach","predicate":"bears_on_proposition","object":{"kind":"entity","entityId":"alignment-is-hard"},"topic":"ideas","statement":"Leike's post of December 2022 argues that developments of recent years have made AI systems more favourable to alignment than they looked initially, and that a more modest goal, aligning a system able to make more alignment progress than humans, is within reach of OpenAI's approach.","time":{"kind":"event","date":"2022-12-05","basis":"publication","note":"Date of posting."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The post argues for a deliberate research programme, not that alignment will be solved by default; it states that a once-and-for-all solution could be very difficult and that success on easy mode might still fail on hard mode.","evidenceIds":["evidence-rd-r06-leike-2022-optimistic-alignment-approach-1","evidence-rd-r06-leike-2022-optimistic-alignment-approach-2","evidence-rd-r06-leike-2022-optimistic-alignment-approach-3"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Argues that aligning a system able to do alignment research is tractable with a deliberate research programme; does not hold that alignment is easy or will be solved by default."}}},{"id":"rd-r07-hawking-russell-tegmark-wilczek-2014-independent","subject":"hawking-russell-tegmark-wilczek-2014-independent","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"ai-extinction-risk"},"topic":"ideas","statement":"The essay by Hawking, Russell, Tegmark and Wilczek of 1 May 2014 states that success in creating AI could be the last event in human history unless the risks are avoided, and that little serious research is devoted to the issue.","time":{"kind":"event","date":"2014-05-01","basis":"publication","note":"Date of publication, as the source gives it."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A short newspaper essay; it does not use the word \"extinction\" and gives no estimate. It records what the four authors wrote, not each author's later views.","evidenceIds":["evidence-rd-r07-hawking-russell-tegmark-wilczek-2014-independent-1","evidence-rd-r07-hawking-russell-tegmark-wilczek-2014-independent-2"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction"}}},{"id":"rd-r12-nature-2023-editorial-stop-talking","subject":"nature-2023-editorial-stop-talking","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"present-harms-priority"},"topic":"ideas","statement":"Nature's unsigned editorial of 27 June 2023 argues that talk of AI destroying humanity draws attention away from the societal harms AI systems are causing now and hinders their regulation.","time":{"kind":"event","date":"2023-06-27","basis":"publication","note":"Date of publication, as the source gives it."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An unsigned editorial; it records the journal's stated position. Statements it attributes to people it spoke with are not quoted.","evidenceIds":["evidence-rd-r12-nature-2023-editorial-stop-talking-1","evidence-rd-r12-nature-2023-editorial-stop-talking-2"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction"}}},{"id":"rd-r14-bender-koller-2020-climbing-nlu","subject":"bender-koller-2020-climbing-nlu","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"llms-lack-agency"},"topic":"ideas","statement":"Bender and Koller's paper of July 2020 argues that a system trained only on linguistic form has a priori no way to learn meaning, so that the language-modelling task cannot in principle lead to learning meaning.","time":{"kind":"event","date":"2020-07","basis":"publication","note":"Date of publication, as the source gives it."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The paper concerns meaning and understanding, defining meaning as the relation between form and communicative intent; it does not discuss whether such systems pursue goals.","evidenceIds":["evidence-rd-r14-bender-koller-2020-climbing-nlu-1","evidence-rd-r14-bender-koller-2020-climbing-nlu-2"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction"}}}],"sources":[{"id":"source-iem","title":"Intelligence Explosion Microeconomics","authors":["Eliezer Yudkowsky"],"publisher":"Machine Intelligence Research Institute","publisherId":"miri","publicationDate":"2013","version":"Technical report 2013-1; last modified September 13, 2013","versionDate":"2013-09-13","canonicalUrl":"https://intelligence.org/files/IEM.pdf","evidenceUrl":"https://intelligence.org/files/IEM.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260917110159/https://intelligence.org/files/IEM.pdf","archivedAt":"2026-09-17","accessedAt":"2026-09-17","reuse":{"attribution":"Machine Intelligence Research Institute, Intelligence Explosion Microeconomics (2013).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and minimal attributed excerpts only. Recheck terms before wider redistribution."}},{"id":"source-anthropic-reinvent-2024","title":"Leverage Anthropic’s Claude models for AI’s evolving landscape","authors":["Maggie Vo","Elie Schoppik"],"publisher":"Anthropic / AWS re:Invent","publicationDate":"2024","version":"AIM123-S; 2024 deck as identified by copyright and event path. PDF page 6 contains founder rows. No exact presentation or upload date inferred.","canonicalUrl":"https://d1.awsstatic.com/onedam/marketing-channels/website/aws/en_US/events/approved/reinvent-2025/reinvent/2024/slides/aim/AIM123-S_Leverage-Anthropics-Claude-models-for-AIs-evolving-landscape-sponsored-by-Anthropic.pdf","evidenceUrl":"https://d1.awsstatic.com/onedam/marketing-channels/website/aws/en_US/events/approved/reinvent-2025/reinvent/2024/slides/aim/AIM123-S_Leverage-Anthropics-Claude-models-for-AIs-evolving-landscape-sponsored-by-Anthropic.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20261029121322/https://d1.awsstatic.com/onedam/marketing-channels/website/aws/en_US/events/approved/reinvent-2025/reinvent/2024/slides/aim/AIM123-S_Leverage-Anthropics-Claude-models-for-AIs-evolving-landscape-sponsored-by-Anthropic.pdf","archivedAt":"2026-10-29","accessedAt":"2026-09-18","reuse":{"attribution":"Maggie Vo; Elie Schoppik, Leverage Anthropic’s Claude models for AI’s evolving landscape (2024).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases or excerpts only. Source roles are relative to each claim."}},{"id":"source-babuschkin-bio","title":"About — Igor Babuschkin","authors":["Igor Babuschkin"],"publisher":"Igor Babuschkin","publicationDate":null,"version":"Undated personal career account, inspected 2026-09-18. Identity cross-checked via github.com/ibab, whose babushk.in link redirects to this domain. Dates and employment titles are not supplied.","canonicalUrl":"https://babuschk.in/about.html","evidenceUrl":"https://babuschk.in/about.html","language":"en","archivedUrl":"https://web.archive.org/web/20260312215654/https://babuschk.in/about.html","archivedAt":"2026-03-12","accessedAt":"2026-09-18","reuse":{"attribution":"Igor Babuschkin, About — Igor Babuschkin.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases or excerpts only. Source roles are relative to each claim."}},{"id":"source-rsp-2-1","documentId":"rsp-v2-1","title":"Responsible Scaling Policy — v2.1","authors":["Anthropic"],"publisher":"Anthropic","publisherId":"anthropic","publicationDate":null,"version":"Responsible Scaling Policy v2.1; effective date 2025-03-31.","versionDate":"2025-03-31","canonicalUrl":"https://www-cdn.anthropic.com/17310f6d70ae5627f55313ed067afc1a762a4068.pdf","evidenceUrl":"https://www-cdn.anthropic.com/17310f6d70ae5627f55313ed067afc1a762a4068.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20250402134435/https://www-cdn.anthropic.com/17310f6d70ae5627f55313ed067afc1a762a4068.pdf","archivedAt":"2025-04-02","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v2.1.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-rsp-3-4","documentId":"rsp-v3-4","title":"Responsible Scaling Policy — v3.4","authors":["Anthropic"],"publisher":"Anthropic","publisherId":"anthropic","publicationDate":null,"version":"Responsible Scaling Policy v3.4; effective date 2026-07-08.","versionDate":"2026-07-08","canonicalUrl":"https://cdn.sanity.io/files/4zrzovbb/website/0bacdc8440ea96e62a8766d99ebe1d4eea6d5f3a.pdf","evidenceUrl":"https://cdn.sanity.io/files/4zrzovbb/website/0bacdc8440ea96e62a8766d99ebe1d4eea6d5f3a.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260710234036/https://cdn.sanity.io/files/4zrzovbb/website/0bacdc8440ea96e62a8766d99ebe1d4eea6d5f3a.pdf","archivedAt":"2026-07-10","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v3.4.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-pf-beta","documentId":"preparedness-beta-2023","title":"Preparedness Framework — Beta","authors":["OpenAI"],"publisher":"OpenAI","publisherId":"openai","publicationDate":"2023-12-18","version":"Preparedness Framework Beta; publication date 2023-12-18.","versionDate":"2023-12-18","canonicalUrl":"https://cdn.openai.com/openai-preparedness-framework-beta.pdf","evidenceUrl":"https://cdn.openai.com/openai-preparedness-framework-beta.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20231218195930/https://cdn.openai.com/openai-preparedness-framework-beta.pdf","archivedAt":"2023-12-18","accessedAt":"2026-09-18","reuse":{"attribution":"OpenAI, Preparedness Framework — Beta.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-gemini-report-v1","documentId":"kp-gemini-family-multimodal-models-2023","title":"Gemini: A Family of Highly Capable Multimodal Models","authors":["Gemini Team, Google"],"publisher":"Gemini Team, Google","publicationDate":"2023-12-19","version":"Pinned arXiv 2312.11805v1, submitted 19 December 2023; section 3 inspected directly. The mutable DeepMind PDF is a later revision and is not the historical evidence.","canonicalUrl":"https://arxiv.org/abs/2312.11805v1","evidenceUrl":"https://arxiv.org/pdf/2312.11805v1","language":"en","accessedAt":"2026-09-19","reuse":{"attribution":"Gemini Team, Google, Gemini: A Family of Highly Capable Multimodal Models (2023-12-19).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and editorial paraphrases only; no full source redistributed."}},{"id":"source-ssi-leadership-2025","title":"Safe Superintelligence Inc. — Updates","authors":["Ilya Sutskever","Safe Superintelligence Inc."],"publisher":"Safe Superintelligence Inc.","publisherId":"ssi","publicationDate":"2025-07-03","version":"Exact July 3, 2025 signed message retained on a mutable multi-date updates page. Later updates are outside these claims.","canonicalUrl":"https://ssi.inc/updates","evidenceUrl":"https://ssi.inc/updates","language":"en","archivedUrl":"https://web.archive.org/web/20260903031538/https://ssi.inc/updates","archivedAt":"2026-09-03","accessedAt":"2026-09-19","reuse":{"attribution":"Ilya Sutskever; Safe Superintelligence Inc., Safe Superintelligence Inc. — Updates (2025-07-03)","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; no full source or personal contact details redistributed."}},{"id":"source-amplification-paper-v1","documentId":"amplification-paper","title":"Supervising strong learners by amplifying weak experts","authors":["Paul Christiano","Buck Shlegeris","Dario Amodei"],"publisher":"arXiv","publicationDate":"2018-10-19","version":"arXiv:1810.08575v1; first-submission date from version history. Later revisions are not substituted.","versionDate":"2018-10-19","canonicalUrl":"https://arxiv.org/abs/1810.08575","evidenceUrl":"https://arxiv.org/pdf/1810.08575v1","language":"en","accessedAt":"2026-09-25","reuse":{"attribution":"Paul Christiano; Buck Shlegeris; Dario Amodei, Supervising strong learners by amplifying weak experts.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; source publications and personal contact details are not redistributed."}},{"id":"source-atlas-openai-lp","title":"OpenAI LP","authors":["Greg Brockman","Ilya Sutskever","OpenAI"],"publisher":"OpenAI","publisherId":"openai","publicationDate":"2019-03-11","version":"Dated publisher page inspected 2026-09-28 and again 2026-10-04 at its current address; the quotes used from 2026-10-04 also match the Internet Archive capture of 2019-03-11 (16:15 UTC) at the original address, which uses straight quotation marks and an additional standfirst.","canonicalUrl":"https://openai.com/index/openai-lp/","evidenceUrl":"https://openai.com/index/openai-lp/","language":"en","archivedUrl":"https://web.archive.org/web/20190311161512/https://openai.com/blog/openai-lp/","archivedAt":"2019-03-11","accessedAt":"2026-10-04","reuse":{"attribution":"Greg Brockman; Ilya Sutskever; OpenAI. OpenAI LP.","licence":"No redistribution licence asserted.","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-cn-r1","title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","authors":["DeepSeek-AI"],"publisher":"DeepSeek-AI","publisherId":"deepseek","publicationDate":"2025-01-22","version":"Source language: English (en). arXiv:2501.12948v1; fixed first version.","canonicalUrl":"https://arxiv.org/html/2501.12948v1","evidenceUrl":"https://arxiv.org/html/2501.12948v1","language":"en","accessedAt":"2026-09-29","reuse":{"attribution":"DeepSeek-AI, DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning (2025-01-22).","licence":"No redistribution licence asserted for this source expression","note":"Bibliographic metadata and brief attributed paraphrases only; model licensing is distinct from source-document reuse."}},{"id":"source-openai-our-structure","title":"Our structure","authors":["OpenAI"],"publisher":"OpenAI","publisherId":"openai","publicationDate":null,"version":"Undated web page describing the structure announced on October 28, 2025; inspected live 2026-10-04. The quoted sentences are identical in the Internet Archive capture of 2025-10-29 (17:22 UTC); the board list differs between the two.","canonicalUrl":"https://openai.com/our-structure/","evidenceUrl":"https://openai.com/our-structure/","language":"en","archivedUrl":"https://web.archive.org/web/20251029172207/https://openai.com/our-structure/","archivedAt":"2025-10-29","accessedAt":"2026-10-04","reuse":{"attribution":"OpenAI, Our structure (web page).","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-anthropic-coi-2023-03-31","documentId":"anthropic-coi-2023-03-31","title":"Amended and Restated Certificate of Incorporation of Anthropic, PBC (filed 2023-03-31)","authors":["Anthropic, PBC"],"publisher":"Anthropic","publisherId":"anthropic","publicationDate":"2023-03-31","version":"Filed 2023-03-31 according to the Division of Corporations stamp on page 1 (file number 4860621); 24-page scan. Delaware-stamped scan published by AI Lab Watch in a shared Google Drive folder (linked from its post of 12 June 2024, which says the filings were obtained from the State of Delaware); downloaded directly on 2026-10-04 and read from page images, as the PDF has no text layer. The Internet Archive holds no capture of this file and no new capture was requested. Authenticity rests on the filing stamp and the consistent recitals of later certificates; no copy was obtained from the Division of Corporations. Publisher as printed on the inspected expression: Anthropic, PBC (filed with the Delaware Division of Corporations).","canonicalUrl":"https://drive.google.com/file/d/1i8Itsv7v7t32lQgzNRqiiO9j5HqSMUVj/view","evidenceUrl":"https://drive.google.com/file/d/1i8Itsv7v7t32lQgzNRqiiO9j5HqSMUVj/view","language":"en","archivedUrl":"https://web.archive.org/web/20261029125140/https://drive.usercontent.google.com/download?id=1i8Itsv7v7t32lQgzNRqiiO9j5HqSMUVj&export=download&authuser=0","archivedAt":"2026-10-29","accessedAt":"2026-10-04","reuse":{"attribution":"Anthropic, PBC, amended and restated certificate of incorporation (filed 2023-03-31), copy published by AI Lab Watch.","licence":"Public filing; no redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-anthropic-coi-2025-07-03","documentId":"anthropic-coi-2025-07-03","title":"Amended and Restated Certificate of Incorporation of Anthropic, PBC (filed 2025-07-03)","authors":["Anthropic, PBC"],"publisher":"Anthropic","publisherId":"anthropic","publicationDate":"2025-07-03","version":"Filed 2025-07-03 according to the Division of Corporations stamp on page 1 (file number 4860621); 26-page scan. Delaware-stamped scan published by AI Lab Watch in a shared Google Drive folder (linked from its post of 12 June 2024, which says the filings were obtained from the State of Delaware); downloaded directly on 2026-10-04 and read from page images, as the PDF has no text layer. The Internet Archive holds no capture of this file and no new capture was requested. Authenticity rests on the filing stamp and the consistent recitals of later certificates; no copy was obtained from the Division of Corporations. Publisher as printed on the inspected expression: Anthropic, PBC (filed with the Delaware Division of Corporations).","canonicalUrl":"https://drive.google.com/file/d/1yeFKDaHN-3o7Zruvd-72De1H1K20Crsh/view","evidenceUrl":"https://drive.google.com/file/d/1yeFKDaHN-3o7Zruvd-72De1H1K20Crsh/view","language":"en","archivedUrl":"https://web.archive.org/web/20261029125714/https://drive.usercontent.google.com/download?id=1yeFKDaHN-3o7Zruvd-72De1H1K20Crsh&export=download","archivedAt":"2026-10-29","accessedAt":"2026-10-04","reuse":{"attribution":"Anthropic, PBC, amended and restated certificate of incorporation (filed 2025-07-03), copy published by AI Lab Watch.","licence":"Public filing; no redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-wh-commitments-2023-07","documentId":"wh-voluntary-ai-commitments-2023-07","title":"Ensuring Safe, Secure, and Trustworthy AI (voluntary AI commitments, July 2023)","authors":["The White House"],"publisher":"The White House","publisherId":"us-white-house","publicationDate":"2023-07-21","version":"PDF linked from the fact sheet of 21 July 2023, inspected 2026-10-04 at its archived-site address; byte-identical (SHA-1) to the Internet Archive captures of 2023-07-21 at the original whitehouse.gov address. The Biden–Harris White House pages are now served from the archived site bidenwhitehouse.archives.gov.","canonicalUrl":"https://bidenwhitehouse.archives.gov/wp-content/uploads/2023/07/Ensuring-Safe-Secure-and-Trustworthy-AI.pdf","evidenceUrl":"https://bidenwhitehouse.archives.gov/wp-content/uploads/2023/07/Ensuring-Safe-Secure-and-Trustworthy-AI.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20230721165220/https://www.whitehouse.gov/wp-content/uploads/2023/07/Ensuring-Safe-Secure-and-Trustworthy-AI.pdf","archivedAt":"2023-07-21","accessedAt":"2026-10-04","reuse":{"attribution":"The White House, Ensuring Safe, Secure, and Trustworthy AI (July 2023).","licence":"US government work; no redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-fasc-2024","documentId":"frontier-ai-safety-commitments-2024","title":"Frontier AI Safety Commitments, AI Seoul Summit 2024","authors":["Department for Science, Innovation and Technology"],"publisher":"Department for Science, Innovation and Technology","publisherId":"dsit","publicationDate":"2024-05-21","version":"GOV.UK HTML publication, first published 21 May 2024 and updated 7 February 2025 to add organisations to the list; inspected live 2026-10-04. The Internet Archive capture of 2024-05-21 (11:05 UTC) has the sixteen-name list only, and the commitment text quoted is identical in both. Publisher as printed on the inspected expression: Department for Science, Innovation and Technology (GOV.UK).","versionDate":"2025-02-07","canonicalUrl":"https://www.gov.uk/government/publications/frontier-ai-safety-commitments-ai-seoul-summit-2024/frontier-ai-safety-commitments-ai-seoul-summit-2024","evidenceUrl":"https://www.gov.uk/government/publications/frontier-ai-safety-commitments-ai-seoul-summit-2024/frontier-ai-safety-commitments-ai-seoul-summit-2024","language":"en","archivedUrl":"https://web.archive.org/web/20240521110530/https://www.gov.uk/government/publications/frontier-ai-safety-commitments-ai-seoul-summit-2024/frontier-ai-safety-commitments-ai-seoul-summit-2024","archivedAt":"2024-05-21","accessedAt":"2026-10-04","reuse":{"attribution":"Department for Science, Innovation and Technology, Frontier AI Safety Commitments, AI Seoul Summit 2024 (GOV.UK).","licence":"Open Government Licence v3.0 (GOV.UK content); quoted briefly under the editorial policy","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-vinge-1983-omni-scan","documentId":"vinge-1983-first-word","title":"First Word (Omni, January 1983, p. 10)","authors":["Vernor Vinge"],"publisher":"Omni","publicationDate":"1983-01","version":"One-page scan hosted by gwern.net, read as an image by the pre-1996 candidate search and a later machine check.","canonicalUrl":"https://gwern.net/doc/transhumanism/1983-vinge.pdf","evidenceUrl":"https://gwern.net/doc/transhumanism/1983-vinge.pdf","language":"en","accessedAt":"2026-10-06","reuse":{"attribution":"Vernor Vinge, \"First Word\", Omni, January 1983, p. 10.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and labelled paraphrases only: the inspected scan is a third-party copy whose permission to host is not established. Phrases in quotation marks inside a paraphrase identify the source's wording."}},{"id":"source-extropians-1998-09-07-yudkowsky-2209","documentId":"extropians-1998-09-07-yudkowsky-2209","title":"Human AI to superhuman (Re: Max More) (Extropians mailing list)","authors":["Eliezer S. Yudkowsky"],"publisher":"Extropians mailing list","publicationDate":"1998-09-07","version":"Archive HTML as served, retrieved 2026-10-06 with the hash recorded by the 1996–2022 candidate search; quoted lines excluded when reading the author’s text. Archive at extropians.weidai.com.","canonicalUrl":"http://extropians.weidai.com/extropians.3Q98/2209.html","evidenceUrl":"http://extropians.weidai.com/extropians.3Q98/2209.html","language":"en","archivedUrl":"https://web.archive.org/web/20150428120647/http://extropians.weidai.com/extropians.3Q98/2209.html","archivedAt":"2015-04-28","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer S. Yudkowsky, \"Human AI to superhuman (Re: Max More)\", Extropians mailing list, 1998-09-07.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-rlo-2019-deceptive-alignment","documentId":"rlo-2019-deceptive-alignment","title":"Deceptive Alignment (Risks from Learned Optimization, part 4)","authors":["Evan Hubinger","Chris van Merwijk","Vladimir Mikulik","Joar Skalse","Scott Garrabrant"],"publisher":"Alignment Forum and LessWrong","publicationDate":"2019-06-05","version":"Text checked in the post’s first stored version through the LessWrong GraphQL API on 2026-10-06.","canonicalUrl":"https://www.alignmentforum.org/posts/zthDPAjh9w6Ytbeks/deceptive-alignment","evidenceUrl":"https://www.alignmentforum.org/posts/zthDPAjh9w6Ytbeks/deceptive-alignment","language":"en","archivedUrl":"https://web.archive.org/web/20190923160252/https://www.alignmentforum.org/posts/zthDPAjh9w6Ytbeks/deceptive-alignment","archivedAt":"2019-09-23","accessedAt":"2026-10-06","reuse":{"attribution":"Evan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, Scott Garrabrant, \"Deceptive Alignment (Risks from Learned Optimization, part 4)\", Alignment Forum and LessWrong, 2019-06-05.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-sl4-2000-07-09-yudkowsky-0007-0017","documentId":"sl4-2000-07-09-yudkowsky-0007-0017","title":"Re: [SL4] Programmed morality (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2000-07-09","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 d84ecf10c4cf385eff5304922fcce6a5fe5f3ef5118efeb6ec597c1c8522e8b2); quoted lines excluded when reading the author’s text. The served archive shows later URL rewriting and does not authenticate historical bytes. Archive at sl4.org.","canonicalUrl":"http://sl4.org/archive/0007/0017.html","evidenceUrl":"http://sl4.org/archive/0007/0017.html","language":"en","archivedUrl":"https://web.archive.org/web/20260216151425/http://sl4.org/archive/0007/0017.html","archivedAt":"2026-02-16","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: [SL4] Programmed morality”, SL4 mailing list, 2000-07-09.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-sl4-2002-06-23-yudkowsky-0206-4219","documentId":"sl4-2002-06-23-yudkowsky-0206-4219","title":"Re: Seed AI (was: How hard a Singularity?) (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2002-06-23","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 8fd31c8bb942dfb3da4906a254fc68c2f4112869ebe5f9811a6f54dc977d7310); quoted lines excluded when reading the author’s text. The served archive shows later URL rewriting and does not authenticate historical bytes. Archive at sl4.org.","canonicalUrl":"http://sl4.org/archive/0206/4219.html","evidenceUrl":"http://sl4.org/archive/0206/4219.html","language":"en","archivedUrl":"https://web.archive.org/web/20260309145702/http://sl4.org/archive/0206/4219.html","archivedAt":"2026-03-09","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: Seed AI (was: How hard a Singularity?)”, SL4 mailing list, 2002-06-23.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-sl4-2002-07-02-yudkowsky-0207-4667","documentId":"sl4-2002-07-02-yudkowsky-0207-4667","title":"Re: Seed AI (was: How hard a Singularity?) (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2002-07-02","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 4e8c20d1200d558941f66899ea4c30e53b1988fef7b7215d35c234c0a5812812); quoted lines excluded when reading the author’s text. The served archive shows later URL rewriting and does not authenticate historical bytes. Archive at sl4.org.","canonicalUrl":"http://sl4.org/archive/0207/4667.html","evidenceUrl":"http://sl4.org/archive/0207/4667.html","language":"en","archivedUrl":"https://web.archive.org/web/20230602125246/http://sl4.org/archive/0207/4667.html","archivedAt":"2023-06-02","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: Seed AI (was: How hard a Singularity?)”, SL4 mailing list, 2002-07-02.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-eu-ai-act-2024","documentId":"eu-ai-act-2024","title":"Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act)","authors":["European Parliament and Council of the European Union"],"publisher":"Official Journal of the European Union (EUR-Lex)","publisherId":"european-parliament-and-council","publicationDate":"2024-07-12","version":"Official Journal L series edition of 12.7.2024, English HTML, retrieved 2026-10-10. The text of the Internet Archive capture of 2024-07-12 matched it apart from the numbering of footnote markers in the amending articles. Later amendments (Regulation (EU) 2026/1744) are not applied to this text.","canonicalUrl":"https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","evidenceUrl":"https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","language":"en","archivedUrl":"https://web.archive.org/web/20240712193938/https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","archivedAt":"2024-07-12","accessedAt":"2026-10-10","reuse":{"attribution":"European Parliament and Council of the European Union, Regulation (EU) 2024/1689 (Artificial Intelligence Act), Official Journal L, 12.7.2024.","licence":"Official legal text; no redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-erdil-besiroglu-2023-explosive-growth-review","documentId":"erdil-besiroglu-2023-explosive-growth-review","title":"Explosive growth from AI automation: A review of the arguments","authors":["Ege Erdil","Tamay Besiroglu"],"publisher":"arXiv:2309.11690 [econ.GN]","publicationDate":"2023-10-01","version":"Wayback capture of arxiv.org/pdf/2309.11690v2; compared with v3. Capture retrieved 2026-10-08 (SHA-256 28610b91d03d1d055d6e1af308b964387905af66557936ee49c622f693985617). Versions: arXiv v1 (2023-09-20): not retrieved; arXiv v2 (2023-10-01): checked; arXiv v3 (2024-07-15): compared. Published version: Erdil, E., & Besiroglu, T. (2023). Explosive growth from AI automation: A review of the arguments (arXiv:2309.11690v2) [Preprint]. arXiv. https://arxiv.org/abs/2309.11690v2 No journal version located (Crossref search 2026-10-10 found none).","canonicalUrl":"http://arxiv.org/pdf/2309.11690v2","evidenceUrl":"https://web.archive.org/web/20250630102426/http://arxiv.org/pdf/2309.11690v2","language":"en","archivedUrl":"https://web.archive.org/web/20250630102426/http://arxiv.org/pdf/2309.11690v2","archivedAt":"2025-06-30","accessedAt":"2026-10-08","reuse":{"attribution":"Ege Erdil, Tamay Besiroglu, “Explosive growth from AI automation: A review of the arguments”, arXiv:2309.11690 [econ.GN], 2023-10-01.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-agrawal-gans-goldfarb-2018-economic-policy-for-ai","documentId":"agrawal-gans-goldfarb-2018-economic-policy-for-ai","title":"Economic Policy for Artificial Intelligence","authors":["Ajay K. Agrawal","Joshua S. Gans","Avi Goldfarb"],"publisher":"NBER Working Paper 24690","publicationDate":"2018-06","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 efbcb773f8553c7db82de09d8a58cadc1836f6cccba250273990531e7d8f7989.","canonicalUrl":"https://www.nber.org/system/files/working_papers/w24690/w24690.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w24690/w24690.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Ajay K. Agrawal, Joshua S. Gans, Avi Goldfarb, “Economic Policy for Artificial Intelligence”, NBER Working Paper 24690, 2018-06.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-acemoglu-restrepo-2019-wrong-kind-of-ai","documentId":"acemoglu-restrepo-2019-wrong-kind-of-ai","title":"The Wrong Kind of AI? Artificial Intelligence and the Future of Labor Demand","authors":["Daron Acemoglu","Pascual Restrepo"],"publisher":"NBER Working Paper 25682","publicationDate":"2019-03","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 54ad101804ab7d30ba6d136f413b1c871ffec0a273dbec290c009acd7ad32100.","canonicalUrl":"https://www.nber.org/system/files/working_papers/w25682/w25682.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w25682/w25682.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Daron Acemoglu, Pascual Restrepo, “The Wrong Kind of AI? Artificial Intelligence and the Future of Labor Demand”, NBER Working Paper 25682, 2019-03.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-diebold-1955-automation-condensation","documentId":"diebold-1955-automation-condensation","title":"A Condensation of the Book Automation, in Automation and Technological Change: Hearings","authors":["John Diebold"],"publisher":"U.S. Congress, Joint Committee on the Economic Report, Subcommittee on Economic Stabilization. (1955). Automation and technological change: Hearings … 84th Congress, 1st session, October 14–28, 1955. Washington: U.S. Government Printing Office, pp. 14–27","publicationDate":"1955-10-14","version":"Internet Archive OCR text (djvu.txt) of a scanned copy, retrieved 2026-10-25 from https://archive.org/download/congress-automation-and-technological-change/congress-automation-and-technological-change_djvu.txt; identical to the copy checked at the 2026-10-14 admission. OCR errors possible; quotes checked only against clean stretches. The cross-family check of 2026-10-25 also inspected the page scans. Current copy; no earlier capture compared. SHA-256 3962809fbb923be62bf986addb0014ee147934a8767475c1f7cfe051959bff90.","canonicalUrl":"https://archive.org/details/congress-automation-and-technological-change","evidenceUrl":"https://archive.org/details/congress-automation-and-technological-change","language":"en","accessedAt":"2026-10-25","reuse":{"attribution":"John Diebold, “A Condensation of the Book Automation, in Automation and Technological Change: Hearings”, U.S. Congress, Joint Committee on the Economic Report, Subcommittee on Economic Stabilization. (1955). Automation and technological change: Hearings … 84th Congress, 1st session, October 14–28, 1955. Washington: U.S. Government Printing Office, pp. 14–27, 1955-10-14.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-leontief-duchin-1984-impacts-of-automation","documentId":"leontief-duchin-1984-impacts-of-automation","title":"The Impacts of Automation on Employment, 1963–2000. Final Report","authors":["Wassily Leontief","Faye Duchin"],"publisher":"New York: Institute for Economic Analysis, New York University; National Science Foundation contract PRA-8012844; April 1984, 376 pp.; ERIC ED 241 743","publicationDate":"1984-04","version":"Internet Archive OCR text (djvu.txt) of a scanned copy, retrieved 2026-10-09 from https://dn711508.ca.archive.org/0/items/microIA411534480031/microIA411534480031djvu.txt; OCR errors possible; quotes checked only against clean stretches. Current copy; no earlier capture compared. SHA-256 cd03df677ca1a723f843eccec0fa1415b3227d604bc83c6783a40321b9a785e9. Book version not retrieved; the 1984 report was. Leontief, W., & Duchin, F. (1986). The future impact of automation on workers. Oxford University Press.","canonicalUrl":"https://archive.org/details/micro_IA41153448_0031","evidenceUrl":"https://archive.org/details/micro_IA41153448_0031","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Wassily Leontief, Faye Duchin, “The Impacts of Automation on Employment, 1963–2000. Final Report”, New York: Institute for Economic Analysis, New York University; National Science Foundation contract PRA-8012844; April 1984, 376 pp.; ERIC ED 241 743, 1984-04.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-cn-zhipu-prospectus-en","title":"Knowledge Atlas Technology Joint Stock Company Limited (北京智譜華章科技股份有限公司) — Global Offering","authors":["Knowledge Atlas Technology Joint Stock Company Limited"],"publisher":"Knowledge Atlas Technology Joint Stock Company Limited (via HKEXnews)","publicationDate":"2025-12-30","version":"Dated document, retrieved on 2026-10-15 (SHA-256 a6598ce3f2f13ac78a9d02a85314dcd349314d2ee1215015e41bb1284e608f4e) directly from the publisher.","canonicalUrl":"https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1230/2025123000017.pdf","evidenceUrl":"https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1230/2025123000017.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20261002230645/https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1230/2025123000017.pdf","archivedAt":"2026-10-02","accessedAt":"2026-10-15","reuse":{"attribution":"Knowledge Atlas Technology Joint Stock Company Limited, “Knowledge Atlas Technology Joint Stock Company Limited (北京智譜華章科技股份有限公司) — Global Offering”, 2025-12-30.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-gw-apple-giannandrea-2025","title":"John Giannandrea to retire from Apple","authors":["Apple"],"publisher":"Apple Newsroom","publicationDate":"2025-12-01","version":"Dated page, retrieved on 2026-10-15 (SHA-256 0c82e3e052f6f538cae379dddb18cb89bf7192569a7027e9117e88bbde6f1b81) directly from the publisher.","canonicalUrl":"https://www.apple.com/newsroom/2025/12/john-giannandrea-to-retire-from-apple/","evidenceUrl":"https://www.apple.com/newsroom/2025/12/john-giannandrea-to-retire-from-apple/","language":"en","archivedUrl":"https://web.archive.org/web/20260921181604/https://www.apple.com/newsroom/2025/12/john-giannandrea-to-retire-from-apple/","archivedAt":"2026-09-21","accessedAt":"2026-10-15","reuse":{"attribution":"Apple, “John Giannandrea to retire from Apple”, 2025-12-01.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-oai-sama-scandinaro-x-2026","title":"Post of 3 February 2026 welcoming OpenAI’s Head of Preparedness","authors":["Sam Altman"],"publisher":"X","publicationDate":"2026-02-03","version":"Dated post, retrieved on 2026-10-15 (SHA-256 079bf097f6d7a89700cb090314238c52f08c7ee40294d7ead61cec6f1062d7a3) through X’s own syndication endpoint for the post’s text, which truncates long notes. Read through X’s syndication endpoint (text only). Publication date is the post’s UTC timestamp.","canonicalUrl":"https://x.com/sama/status/2018813527780463027","evidenceUrl":"https://x.com/sama/status/2018813527780463027","language":"en","archivedUrl":"https://web.archive.org/web/20260203222707/https://twitter.com/sama/status/2018813527780463027","archivedAt":"2026-02-03","accessedAt":"2026-10-15","reuse":{"attribution":"Sam Altman, “Post of 3 February 2026 welcoming OpenAI’s Head of Preparedness”, 2026-02-03.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-oai-scandinaro-x-2026","title":"Post of 3 February 2026 on joining OpenAI as Head of Preparedness","authors":["Dylan Scandinaro"],"publisher":"X","publicationDate":"2026-02-03","version":"Dated post, retrieved on 2026-10-15 (SHA-256 92f86feb73392411756565f98f3f31233b7083009536a628f2ae9a9b65ed4331) through X’s own syndication endpoint for the post’s text, which truncates long notes. Read through X’s syndication endpoint (text only). Publication date is the post’s UTC timestamp.","canonicalUrl":"https://x.com/dylanscand/status/2018805998350455143","evidenceUrl":"https://x.com/dylanscand/status/2018805998350455143","language":"en","archivedUrl":"https://web.archive.org/web/20260203215712/https://twitter.com/dylanscand/status/2018805998350455143","archivedAt":"2026-02-03","accessedAt":"2026-10-15","reuse":{"attribution":"Dylan Scandinaro, “Post of 3 February 2026 on joining OpenAI as Head of Preparedness”, 2026-02-03.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-oai-ssc-2024","title":"OpenAI Board Forms Safety and Security Committee","authors":["OpenAI"],"publisher":"OpenAI","publisherId":"openai","publicationDate":"2024-05-28","version":"Dated page, retrieved on 2026-10-15 (SHA-256 34a351ce00b93ca65fdcd73f59225314098c44da2c1865630989fa68cd213a52) through the third-party reader relay r.jina.ai, which renders the publisher’s current page (the publisher refused direct retrieval); not an archive copy. The inspected text carries an update of 18 June 2024 above the original announcement.","canonicalUrl":"https://openai.com/index/openai-board-forms-safety-and-security-committee/","evidenceUrl":"https://openai.com/index/openai-board-forms-safety-and-security-committee/","language":"en","archivedUrl":"https://web.archive.org/web/20260713143403/https://openai.com/index/openai-board-forms-safety-and-security-committee/","archivedAt":"2026-07-13","accessedAt":"2026-10-15","reuse":{"attribution":"OpenAI, “OpenAI Board Forms Safety and Security Committee”, 2024-05-28.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"c2-src-baidu-20f-fy2023","title":"Baidu, Inc. annual report on Form 20-F for the fiscal year ended December 31, 2023","authors":["Baidu, Inc."],"publisher":"U.S. Securities and Exchange Commission (EDGAR)","publicationDate":"2024-03-15","version":"Dated document, retrieved on 2026-10-27 (SHA-256 745a5397f9149a6d957a64226a03cc382735a41e5b81772786daa538ddd882a3) through the third-party reader relay r.jina.ai, which renders the publisher’s current page (direct retrieval failed); not an archive copy. Version: 0001193125-24-068527. Filed document d584913d20f.htm, filing date 15 March 2024; retrieved through the r.jina.ai relay after direct retrieval failed (HTTP/2 stream errors).","canonicalUrl":"https://www.sec.gov/Archives/edgar/data/1329099/000119312524068527/d584913d20f.htm","evidenceUrl":"https://www.sec.gov/Archives/edgar/data/1329099/000119312524068527/d584913d20f.htm","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"Baidu, Inc., “Baidu, Inc. annual report on Form 20-F for the fiscal year ended December 31, 2023”, 2024-03-15.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"c2-src-ernie5-arxiv-v1","title":"ERNIE 5.0 Technical Report","authors":["ERNIE Team, Baidu"],"publisher":"arXiv","publicationDate":"2026-02-04","version":"Dated document, retrieved on 2026-10-27 (SHA-256 a1450a354beb2e94f871c41d230c11c03f319be07dd665f6f6524cc6864abaee) directly from the publisher. Version: 2602.04705v1. Version 1 as submitted on 4 February 2026.","canonicalUrl":"https://arxiv.org/pdf/2602.04705v1","evidenceUrl":"https://arxiv.org/pdf/2602.04705v1","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"ERNIE Team, Baidu, “ERNIE 5.0 Technical Report”, 2026-02-04.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"c2-src-tc260-asgf-v2-pdf","documentId":"tc260-asgf-v2-0","title":"人工智能安全治理框架2.0","authors":["全国网络安全标准化技术委员会","国家计算机网络应急技术处理协调中心"],"publisher":"全国网络安全标准化技术委员会","publisherId":"tc260","publicationDate":"2025-09","version":"Dated document, retrieved on 2026-10-27 (SHA-256 50735b9ca01bff746169fa58e554abb97b4e62c27d4c97afd64f18ff83bba705) directly from the publisher. Version: 1759653474200838.pdf. Bilingual PDF (Chinese text followed by the publisher’s English text) linked from the CAC news item of 15 September 2025. Quotes are from the Chinese text, transcribed as printed: the text layer contains control characters (U+0001) and compatibility forms (Kangxi radicals, ⻛ for 风, ⻅ for 见), which are normalised in the quotations. No copy on tc260.org.cn was located. English title (machine translation): “AI Safety Governance Framework 2.0”. Publisher as printed on the inspected expression: 全国网络安全标准化技术委员会 (hosted by the Cyberspace Administration of China, cac.gov.cn).","canonicalUrl":"https://www.cac.gov.cn/rootimages/uploadimg/1759653474200838/1759653474200838.pdf","evidenceUrl":"https://www.cac.gov.cn/rootimages/uploadimg/1759653474200838/1759653474200838.pdf","language":"zh-Hans","archivedUrl":"https://web.archive.org/web/20260302034821/https://www.cac.gov.cn/rootimages/uploadimg/1759653474200838/1759653474200838.pdf","archivedAt":"2026-03-02","accessedAt":"2026-10-27","reuse":{"attribution":"全国网络安全标准化技术委员会, 国家计算机网络应急技术处理协调中心, “人工智能安全治理框架2.0”, 2025-09.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-uk-cohere-mou-2025","documentId":"uk-cohere-mou-2025","title":"Memorandum of Understanding between UK and Cohere on AI opportunities","authors":["Department for Science, Innovation and Technology"],"publisher":"Department for Science, Innovation and Technology","publisherId":"dsit","publicationDate":"2025-06-16","version":"Retrieved on 2026-10-27 (SHA-256 a1ec5955406d1e893746f38e912b8a42056050eedc2319f9f3bcabd81a577ca2). GOV.UK HTML text, 'Published 16 June 2025'; no signatory block or signing date in the published text. Publisher as printed on the inspected expression: GOV.UK (Department for Science, Innovation and Technology).","canonicalUrl":"https://www.gov.uk/government/publications/memorandum-of-understanding-between-the-uk-and-cohere-on-ai-opportunities/memorandum-of-understanding-between-uk-and-cohere-on-ai-opportunities","evidenceUrl":"https://www.gov.uk/government/publications/memorandum-of-understanding-between-the-uk-and-cohere-on-ai-opportunities/memorandum-of-understanding-between-uk-and-cohere-on-ai-opportunities","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"Department for Science, Innovation and Technology, “Memorandum of Understanding between UK and Cohere on AI opportunities”, 2025-06-16.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cpo-pw-eo-14299-fr","documentId":"us-eo-14299-2025","title":"Executive Order 14299: Deploying Advanced Nuclear Reactor Technologies for National Security (90 FR 22581)","authors":["The White House"],"publisher":"Federal Register (Office of the Federal Register, National Archives) via govinfo.gov","publisherId":"us-white-house","publicationDate":"2025-05-29","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 22ddba09a2e933ddc8cf0c82c11641358f620da7566bc693aac46e6a6434a4b7). Federal Register print of 29 May 2025, Vol. 90, No. 102, pp. 22581 ff., FR Doc. 2025-09796. The PDF text layer scatters the title words. Retrieved again at admission on 2026-10-30 (SHA-256 22ddba09a2e933ddc8cf0c82c11641358f620da7566bc693aac46e6a6434a4b7, the same bytes).","canonicalUrl":"https://www.govinfo.gov/content/pkg/FR-2025-05-29/pdf/2025-09796.pdf","evidenceUrl":"https://www.govinfo.gov/content/pkg/FR-2025-05-29/pdf/2025-09796.pdf","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"The White House, “Executive Order 14299: Deploying Advanced Nuclear Reactor Technologies for National Security (90 FR 22581)”, 2025-05-29.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cpo-pw-constellation-google-890-2026","title":"Google and Constellation Announce Landmark Agreement to Bring 890 MW of New Nuclear Capacity to PJM Grid as Part of Long-Term Power Deal","authors":["Constellation Energy"],"publisher":"Constellation Energy","publicationDate":"2026-10-06","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 760f30f2210ec55e54700b3298b48ab64205545e4c61d612c337dd18dd492c56). Joint release page on constellationenergy.com as served on 2026-10-28; dateline BALTIMORE and MOUNTAIN VIEW, Calif. — Oct. 6, 2026. Google's own post on the agreement was not retrieved. Retrieved again at admission on 2026-10-30 (SHA-256 760f30f2210ec55e54700b3298b48ab64205545e4c61d612c337dd18dd492c56, the same bytes). Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://www.constellationenergy.com/news/2026/10/google-and-constellation-announce-landmark-agreement-to-bring-890-mw-of-new-nuclear-capacity-to-pjm-grid.html","evidenceUrl":"https://www.constellationenergy.com/news/2026/10/google-and-constellation-announce-landmark-agreement-to-bring-890-mw-of-new-nuclear-capacity-to-pjm-grid.html","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Constellation Energy, “Google and Constellation Announce Landmark Agreement to Bring 890 MW of New Nuclear Capacity to PJM Grid as Part of Long-Term Power Deal”, 2026-10-06.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-aib-esm1b-paper","documentId":"aib-doc-esm1b-2021","title":"Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences","authors":["Alexander Rives","Joshua Meier","Tom Sercu","Siddharth Goyal","Zeming Lin","Jason Liu","Demi Guo","Myle Ott","C. Lawrence Zitnick","Jerry Ma","Rob Fergus"],"publisher":"Proceedings of the National Academy of Sciences (open access copy in PubMed Central)","publicationDate":"2021-04-05","version":"Retrieved for the candidate pack on 2026-10-30 (SHA-256 5a365c9b9bca1d0cb6065822636366ef838a8192d09787229202ea2c5628b236). PMC8053943 copy of PNAS 118(15):e2016239118.","canonicalUrl":"https://pmc.ncbi.nlm.nih.gov/articles/PMC8053943/","evidenceUrl":"https://pmc.ncbi.nlm.nih.gov/articles/PMC8053943/","language":"en","accessedAt":"2026-10-30","reuse":{"attribution":"Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, Rob Fergus, “Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences”, 2021-04-05.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-kp-opportunities-risks-foundation-models-2021-v1","documentId":"kp-opportunities-risks-foundation-models-2021","title":"On the Opportunities and Risks of Foundation Models","authors":["Rishi Bommasani","Drew A. Hudson","Ehsan Adeli","Russ Altman","Simran Arora","Sydney von Arx","Michael S. Bernstein","Jeannette Bohg","Antoine Bosselut","Emma Brunskill","Erik Brynjolfsson","Shyamal Buch","Dallas Card","Rodrigo Castellon","Niladri Chatterji","Annie Chen","Kathleen Creel","Chris Donahue","Moussa Doumbouya","Esin Durmus","Stefano Ermon","John Etchemendy","Kawin Ethayarajh","Chelsea Finn","Trevor Gale","Lauren Gillespie","Karan Goel","Noah Goodman","Shelby Grossman","Neel Guha","Tatsunori Hashimoto","Peter Henderson","John Hewitt","Daniel E. Ho","Jenny Hong","Kyle Hsu","Jing Huang","Thomas Icard","Saahil Jain","Dan Jurafsky","Pratyusha Kalluri","Siddharth Karamcheti","Geoff Keeling","Fereshte Khani","Omar Khattab","Pang Wei Koh","Mark Krass","Ranjay Krishna","Rohith Kuditipudi","Ananya Kumar","Faisal Ladhak","Mina Lee","Tony Lee","Jure Leskovec","Isabelle Levent","Xiang Lisa Li","Xuechen Li","Tengyu Ma","Ali Malik","Christopher D. Manning","Suvir Mirchandani","Eric Mitchell","Zanele Munyikwa","Suraj Nair","Avanika Narayan","Deepak Narayanan","Allen Nie","Juan Carlos Niebles","Hamed Nilforoshan","Julian Nyarko","Giray Ogut","Laurel Orr","Isabel Papadimitriou","Chris Piech","Eva Portelance","Christopher Potts","Aditi Raghunathan","Hongyu Ren","Frieda Rong","Yusuf Roohani","Camilo Ruiz","Jack Ryan","Christopher Ré","Dorsa Sadigh","Shiori Sagawa","Keshav Santhanam","Andy Shih","Krishnan Srinivasan","Alex Tamkin","Rohan Taori","Armin W. Thomas","Florian Tramèr","Rose E. Wang","William Wang","Bohan Wu","Jiajun Wu","Yuhuai Wu","Sang Michael Xie","Michihiro Yasunaga","Jiaxuan You","Matei Zaharia","Michael Zhang","Tianyi Zhang","Xikun Zhang","Yuhui Zhang","Lucia Zheng","Kaitlyn Zhou","Percy Liang"],"publisher":"arXiv","publicationDate":"2021-08-16","version":"arXiv:2108.07258v1, submitted 16 August 2021 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 7d7de90261ec22654cd360b7925886d75321db35ab6ff9412d8893da760ffe67). The author list is the byline as printed on the first page.","versionDate":"2021-08-16","canonicalUrl":"https://arxiv.org/abs/2108.07258v1","evidenceUrl":"https://arxiv.org/pdf/2108.07258v1","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Rishi Bommasani and others, “On the Opportunities and Risks of Foundation Models”, 2021-08-16.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-kp-mmlu-pro-more-robust-challenging-2024-v1","documentId":"kp-mmlu-pro-more-robust-challenging-2024","title":"MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark","authors":["Yubo Wang","Xueguang Ma","Ge Zhang","Yuansheng Ni","Abhranil Chandra","Shiguang Guo","Weiming Ren","Aaran Arulraj","Xuan He","Ziyan Jiang","Tianle Li","Max Ku","Kai Wang","Alex Zhuang","Rongqi Fan","Xiang Yue","Wenhu Chen"],"publisher":"arXiv","publicationDate":"2024-06-03","version":"arXiv:2406.01574v1, submitted 3 June 2024 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 9bf7b08bf1c220cb6c1ed0dcafb7d10848a21178137e13dd5e13981d8ecd15e3). The author list is the byline as printed on the first page.","versionDate":"2024-06-03","canonicalUrl":"https://arxiv.org/abs/2406.01574v1","evidenceUrl":"https://arxiv.org/pdf/2406.01574v1","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Yubo Wang and others, “MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark”, 2024-06-03.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-kp-adversarial-winograd-schema-challenge-at-2019-v1","documentId":"kp-adversarial-winograd-schema-challenge-at-2019","title":"WinoGrande: An Adversarial Winograd Schema Challenge at Scale","authors":["Keisuke Sakaguchi","Ronan Le Bras","Chandra Bhagavatula","Yejin Choi"],"publisher":"arXiv","publicationDate":"2019-07-24","version":"arXiv:1907.10641v1, submitted 24 July 2019 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 e97b9fa0b7d8aaa9c17cbb954cef105d2c2040d40b23bda37891fecab6621427). The author list is the byline as printed on the first page.","versionDate":"2019-07-24","canonicalUrl":"https://arxiv.org/abs/1907.10641v1","evidenceUrl":"https://arxiv.org/pdf/1907.10641v1","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Keisuke Sakaguchi, Ronan Le Bras, Chandra Bhagavatula, Yejin Choi, “WinoGrande: An Adversarial Winograd Schema Challenge at Scale”, 2019-07-24.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-kp-mastering-game-go-deep-neural-2016-text","documentId":"kp-mastering-game-go-deep-neural-2016","title":"Mastering the game of Go with deep neural networks and tree search","authors":["David Silver","Aja Huang","Chris J. Maddison","Arthur Guez","Laurent Sifre","Julian Schrittwieser","Ioannis Antonoglou","Veda Panneershelvam","Marc Lanctot","Sander Dieleman","Dominik Grewe","John Nham","Nal Kalchbrenner","Ilya Sutskever","Timothy Lillicrap","Madeleine Leach","Koray Kavukcuoglu","Thore Graepel","Demis Hassabis"],"publisher":"Nature","publicationDate":"2016","version":"Publisher’s article page (abstract only). Year from Crossref. Retrieved for the candidate pack on 2026-10-28 (SHA-256 e537f7be8a95a8c3b15d3c8ee62b92976b9a86d14c430fa71c35dd349fb62f8d). The author list is the byline as printed on the first page.","canonicalUrl":"https://www.nature.com/articles/nature16961","evidenceUrl":"https://www.nature.com/articles/nature16961","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"David Silver and others, “Mastering the game of Go with deep neural networks and tree search”, 2016.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"c3-src-iflytek-prospectus-2008","title":"安徽科大讯飞信息科技股份有限公司首次公开发行股票招股说明书","authors":["安徽科大讯飞信息科技股份有限公司"],"publisher":"Anhui USTC iFlytek Co., Ltd. (via cninfo)","publicationDate":"2008-04-24","version":"Dated document, retrieved on 2026-10-29 (SHA-256 2e56e7abfd8bd36eccbaabf81701311cc64c5d28ac2aabe0b45fdf92fc0a6c72) directly from the publisher. Version: cninfo finalpage/2008-04-24/39107260.PDF. IPO prospectus as published on cninfo on 24 April 2008. English title (machine translation): “Anhui USTC iFlytek Co., Ltd. — Prospectus for the initial public offering of shares”.","canonicalUrl":"https://static.cninfo.com.cn/finalpage/2008-04-24/39107260.PDF","evidenceUrl":"https://static.cninfo.com.cn/finalpage/2008-04-24/39107260.PDF","language":"zh-Hans","accessedAt":"2026-10-29","reuse":{"attribution":"安徽科大讯飞信息科技股份有限公司, “安徽科大讯飞信息科技股份有限公司首次公开发行股票招股说明书”, 2008-04-24.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-rd-bender-koller-2020-climbing-nlu","documentId":"bender-koller-2020-climbing-nlu","title":"Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data","authors":["Emily M. Bender","Alexander Koller"],"publisher":"Proceedings of the 58th Annual Meeting of the ACL (ACL Anthology 2020.acl-main.463)","publicationDate":"2020-07","version":"Retrieved for the candidate pack on 2026-10-29 (SHA-256 48823a487ee233ff567aa3fdd3c3f20f8425f0182b6f213c4c7f1e929d2bff77). Published proceedings PDF, ACL Anthology.","canonicalUrl":"https://aclanthology.org/2020.acl-main.463.pdf","evidenceUrl":"https://aclanthology.org/2020.acl-main.463.pdf","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"Emily M. Bender, Alexander Koller, “Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data”, 2020-07.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-rd-hawking-russell-tegmark-wilczek-2014-independent","documentId":"hawking-russell-tegmark-wilczek-2014-independent","title":"Stephen Hawking: 'Transcendence looks at the implications of artificial intelligence - but are we taking AI seriously enough?'","authors":["Stephen Hawking","Stuart Russell","Max Tegmark","Frank Wilczek"],"publisher":"The Independent (online)","publicationDate":"2014-05-01","version":"Retrieved for the candidate pack on 2026-10-29 (SHA-256 ce50f0b1dbd8d440c7b21fc49d050feb8d50e47a3a253475a76d2f5a978a831b). Current online article, dated Thursday 01 May 2014 21:30 BST on the page; no update note.","canonicalUrl":"https://www.independent.co.uk/news/science/stephen-hawking-transcendence-looks-at-the-implications-of-artificial-intelligence-but-are-we-taking-ai-seriously-enough-9313474.html","evidenceUrl":"https://www.independent.co.uk/news/science/stephen-hawking-transcendence-looks-at-the-implications-of-artificial-intelligence-but-are-we-taking-ai-seriously-enough-9313474.html","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"Stephen Hawking, Stuart Russell, Max Tegmark, Frank Wilczek, “Stephen Hawking: 'Transcendence looks at the implications of artificial intelligence - but are we taking AI seriously enough?'”, 2014-05-01.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-rd-leike-2022-optimistic-alignment-approach","documentId":"leike-2022-optimistic-alignment-approach","title":"Why I'm optimistic about our alignment approach","authors":["Jan Leike"],"publisher":"Musings on the Alignment Problem (aligned.substack.com), the author's newsletter","publicationDate":"2022-12-05","version":"Retrieved for the candidate pack on 2026-10-29 (SHA-256 2717984f32c8817f752d65e183dc8bf8ab6bdf286d7fcc7c698e17aca78569e0). Current page as served on 2026-10-29; dated 5 December 2022 on the page.","canonicalUrl":"https://aligned.substack.com/p/alignment-optimism","evidenceUrl":"https://aligned.substack.com/p/alignment-optimism","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"Jan Leike, “Why I'm optimistic about our alignment approach”, 2022-12-05.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-rd-nature-2023-editorial-stop-talking","documentId":"nature-2023-editorial-stop-talking","title":"Stop talking about tomorrow's AI doomsday when AI poses risks today","authors":["Nature"],"publisher":"Nature 618, 885–886 (nature.com)","publicationDate":"2023-06-27","version":"Retrieved for the candidate pack on 2026-10-29 (SHA-256 77439706f00e59db2b7f32d09e1c1981b36a67b1b2385c2d04b3f8400310d747). Current online editorial, dated 27 June 2023 in its metadata.","canonicalUrl":"https://www.nature.com/articles/d41586-023-02094-7?error=cookies_not_supported&code=8a8ae576-f949-430d-96e2-424c182aa80a","evidenceUrl":"https://www.nature.com/articles/d41586-023-02094-7?error=cookies_not_supported&code=8a8ae576-f949-430d-96e2-424c182aa80a","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"Nature, “Stop talking about tomorrow's AI doomsday when AI poses risks today”, 2023-06-27.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}}],"evidence":[{"id":"evidence-babuschkin-openai-prior","claimId":"babuschkin-openai-prior","sourceId":"source-babuschkin-bio","sourceRole":"primary","support":"supports","locator":"The Journey, OpenAI paragraph","passage":"His account describes OpenAI work on reasoning and pretraining.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-18","note":"Relevant source passage inspected by an automated research assistant. Human source review is pending."},{"id":"evidence-rsp-v2-1-rd-thresholds","claimId":"rsp-v2-1-rd-thresholds","sourceId":"source-rsp-2-1","sourceRole":"primary","support":"supports","locator":"PDF p. 22 (printed p. 18), Changelog, March 31, 2025 (RSP v2.1) entry, paragraph \"New Capability Thresholds\", second sentence","passage":"We have also disaggregated our existing AI R&D capability thresholds, separating them into two distinct levels (the ability to fully automate entry-level AI research work, and the ability to cause dramatic acceleration in the rate of effective scaling)","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical to the Internet Archive capture of 2025-04-02; it replaces an earlier attributed paraphrase. Human editorial review is pending."},{"id":"evidence-sutskever-ssi-ceo-2025","claimId":"sutskever-ssi-ceo-2025","sourceId":"source-ssi-leadership-2025","sourceRole":"primary","support":"supports","locator":"July 3rd, 2025 update, second paragraph and Ilya signature","passage":"The signed message states that Sutskever is formally CEO.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-19","note":"Source passage independently inspected by an automated research assistant for this scoped record. Human source review pending."},{"id":"evidence-cn-r1-rl","claimId":"cn-r1-rl","sourceId":"source-cn-r1","sourceRole":"primary","support":"supports","locator":"§2.1 Overview; §2.2 and §2.3","passage":"R1-Zero applies reinforcement learning directly; R1 begins from a checkpoint fine-tuned on cold-start examples.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-29","note":"English source paraphrased and checked by machine. Exploratory seed, not editorially accepted. Source reporting is not independent verification of implementation."},{"id":"evidence-prov-rsp-v3-4-competitor-lead-delay","claimId":"prov-rsp-v3-4-competitor-lead-delay","sourceId":"source-rsp-3-4","sourceRole":"primary","support":"supports","locator":"PDF p. 17 (printed p. 17), Appendix A: Commitments Related to Competitors, table row 'Anthropic in the lead', Commitment column","passage":"We will require a strong argument that catastrophic risk is contained, along the lines of our recommendations for industry-wide safety (see Section 1). We will delay AI development and deployment as needed to achieve this, until and unless we no longer believe we have a significant lead.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical to the Internet Archive capture of 2026-07-10. Extraction note: Unchanged from v3.3. Human editorial review is pending."},{"id":"evidence-cs-anthropic-ltbt-removes-directors-1","claimId":"cs-anthropic-ltbt-removes-directors","sourceId":"source-anthropic-coi-2025-07-03","sourceRole":"primary","support":"supports","locator":"Article VIII(B)7 “Director Removal” (page 23)","passage":"Any director elected as provided in Article VIII(B) may be removed at any time without cause by the affirmative vote of the holders of the shares of the class or series of capital stock entitled to elect such director or directors","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-04","note":"Verbatim quote transcribed from the page image by the integrating session on 2026-10-04 and compared with the scan; the copy has no reliable text layer, so the match is visual, not mechanical. The integrating session wrote or selected this passage from its own reading of the source; it does not repeat a passage from the week 4b research pass. Human editorial review is pending."},{"id":"evidence-cs-anthropic-ltbt-removes-directors-2","claimId":"cs-anthropic-ltbt-removes-directors","sourceId":"source-anthropic-coi-2023-03-31","sourceRole":"primary","support":"context","locator":"Article VIII(B)7 (page 21)","passage":"The 2023 certificate contains the same director-removal provision.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-04","note":"Editorial paraphrase of the page image, re-checked on 2026-10-04 by the integrating session against the scan. The integrating session wrote or selected this passage from its own reading of the source; it does not repeat a passage from the week 4b research pass. Human editorial review is pending."},{"id":"evidence-prov-wh-2023-07-public-reporting-1","claimId":"prov-wh-2023-07-public-reporting","sourceId":"source-wh-commitments-2023-07","sourceRole":"primary","support":"supports","locator":"Page 4, commitment 6","passage":"publish reports for all new significant model public releases within scope","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-04","note":"Verbatim quote, re-checked on 2026-10-04 by the integrating session against its own direct fetch of the cited expression (mechanical match after whitespace normalisation). The passage was first proposed and machine-verified in the week 4b research pass. Human editorial review is pending."},{"id":"evidence-cm-seoul-xai-1","claimId":"cm-seoul-xai","sourceId":"source-fasc-2024","sourceRole":"primary","support":"supports","locator":"Opening sentence and first list of organisations (sixteen names)","passage":"The text opens by stating that the UK and Republic of Korea governments announced that the listed organisations have agreed to the commitments, and the list includes “xAI”.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-04","note":"Editorial paraphrase, re-checked on 2026-10-04 by the integrating session against its own direct fetch of the cited expression. The list is identical in the Internet Archive capture of 2024-05-21. The integrating session wrote or selected this passage from its own reading of the source; it does not repeat a passage from the week 4b research pass. Human editorial review is pending."},{"id":"evidence-ts-b6-vinge-1983-context","claimId":"ts-b6-vinge-1983-context","sourceId":"source-vinge-1983-omni-scan","sourceRole":"primary","support":"supports","locator":"Omni p. 10, right column, on successors and human leadership","passage":"Vinge expects human leadership to end, and considers human-computer partnership as an alternative form of amplification.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-06","note":"Paraphrase from the pre-1996 candidate search."},{"id":"evidence-yk-t08-s1","claimId":"yk-y08-sl4-2000-07-09-yudkowsky-0007-0017-b","sourceId":"source-sl4-2000-07-09-yudkowsky-0007-0017","sourceRole":"primary","support":"supports","locator":"message body, paragraph beginning 'I do think a general Web connection would be a good thing'","passage":"By the time we've actually worked with AI long enough for it to be anything remotely like a threat, we'll have an excellent mental model of what goes on inside the AI's motivations and we won't feel all that much nervousness.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-07","note":"Candidate pack passage T08-s1. The quotation (41 words) matched the retrieved expression (SHA-256 d84ecf10c4cf385eff5304922fcce6a5fe5f3ef5118efeb6ec597c1c8522e8b2) when checked on 2026-10-07.","contextOnly":{"reason":"An earlier view that only partly matches the take; never show alone","requires":"His later text rejecting extrapolation from observed good behaviour, which does not refer back to this message.","claimIds":["yk-y08-sl4-2006-06-06-yudkowsky-0606-15174-c"]}},{"id":"evidence-yk-t08-s5","claimId":"yk-y08-sl4-2006-06-06-yudkowsky-0606-15174-c","sourceId":"source-sl4-2006-06-06-yudkowsky-0606-15174","passage":"I can think of *many* failure modes such that the AI appears to behave well as a prehuman, then wipes out humanity as a posthuman.","passageType":"quote"},{"id":"evidence-yk-t13-s3","claimId":"yk-y13-sl4-2002-07-02-yudkowsky-0207-4667-b","sourceId":"source-sl4-2002-07-02-yudkowsky-0207-4667","sourceRole":"primary","support":"supports","locator":"message body, paragraph beginning 'Necessarily? Only if you value the human species.'","passage":"you would be hellishly late in your Friendship work if you tried to get started on it then.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-07","note":"Candidate pack passage T13-s3. The quotation (18 words) matched the retrieved expression (SHA-256 4e8c20d1200d558941f66899ea4c30e53b1988fef7b7215d35c234c0a5812812) when checked on 2026-10-07.","contextOnly":{"reason":"About development stage, not timelines; never show as a timeline claim","requires":"The opening of the sentence, given in the qualification, which ties the point to the stage of the AI's development, not to calendar time."}},{"id":"evidence-ec-g09-f","claimId":"ec-g09-erdil-besiroglu-2023-explosive-growth-review-a","sourceId":"source-erdil-besiroglu-2023-explosive-growth-review","sourceRole":"primary","support":"supports","locator":"Section 1 (Introduction), fourth theme","passage":"We think that the odds of widespread automation and subsequent explosive growth by the end of this century are about even.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage G09-f (research/economics-growth). The quotation (21 words) matched the checked expression (erdil-arxiv-v2) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m01-11","claimId":"ec-m01-leontief-duchin-1984-impacts-of-automation-b","sourceId":"source-leontief-duchin-1984-impacts-of-automation","sourceRole":"primary","support":"supports","locator":"printed p. 1.33","passage":"The electronic revolution became visible only a few years ago, and by the year 2000 it will be not more advanced than the mechanization of European economies had advanced by, say, the year 1820.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M01-11 (research/economics-mainstream). The quotation (34 words) matched the retrieved current copy (SHA-256 cd03df677ca1a723f843eccec0fa1415b3227d604bc83c6783a40321b9a785e9) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day.","contextOnly":{"reason":"Fresh-context audit (A2) judged that the quotation misleads when shown alone.","requires":"The preceding sentences state that major break-throughs (automatic programming, speech recognition, robot vision) are explicitly EXCLUDED from the scenarios."}},{"id":"evidence-ec-m09-23","claimId":"ec-m09-agrawal-gans-goldfarb-2018-economic-policy-for-ai-b","sourceId":"source-agrawal-gans-goldfarb-2018-economic-policy-for-ai","sourceRole":"primary","support":"supports","locator":"pdf page 18; section 3 (AI and jobs)","passage":"In summary, the near consensus among economists is optimism for the long run prospects of jobs and growth; however, the long run can be a very long time.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M09-23 (research/economics-mainstream). The quotation (28 words) matched the retrieved current copy (SHA-256 efbcb773f8553c7db82de09d8a58cadc1836f6cccba250273990531e7d8f7989) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"lg-ev-oai-ssc-formed-2024","claimId":"lg-oai-ssc-formed-2024","sourceId":"lg-src-oai-ssc-2024","sourceRole":"primary","support":"supports","locator":"Paragraph beginning “Today, the OpenAI Board formed”","passage":"The announcement says the OpenAI Board formed the committee to make recommendations to the full Board on critical safety and security decisions.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “Board formed a Safety and Security Committee” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through the third-party reader relay r.jina.ai, which renders the publisher’s current page (the publisher refused direct retrieval); not an archive copy. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. No human editorial review.","retrievalRoute":"reader_relay"},{"id":"lg-ev-oai-ssc-formed-2024-inc","claimId":"lg-oai-ssc-formed-2024","sourceId":"source-atlas-openai-lp","sourceRole":"primary","support":"context","locator":"Paragraph defining the general partner","passage":"OpenAI’s 2019 note gives the legal name of OpenAI Nonprofit as OpenAI Inc.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “whose legal name is “OpenAI Inc”” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through the third-party reader relay r.jina.ai, which renders the publisher’s current page (the publisher refused direct retrieval); not an archive copy. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. No human editorial review.","retrievalRoute":"reader_relay"},{"id":"lg-ev-oai-ssc-formed-2024-governed","claimId":"lg-oai-ssc-formed-2024","sourceId":"source-openai-our-structure","sourceRole":"primary","support":"context","locator":"Opening section (page retrieved on 15 October 2026)","passage":"The structure page states that the for-profit created in 2019 has always been governed and controlled by the nonprofit, now the OpenAI Foundation.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “governed and controlled by the nonprofit” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through the third-party reader relay r.jina.ai, which renders the publisher’s current page (the publisher refused direct retrieval); not an archive copy. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. No human editorial review.","retrievalRoute":"reader_relay"},{"id":"lg-ev-oai-scandinaro-preparedness-2026-1","claimId":"lg-oai-scandinaro-preparedness-2026","sourceId":"lg-src-oai-sama-scandinaro-x-2026","sourceRole":"primary","support":"supports","locator":"Post text, first sentence","passage":"Altman welcomes Scandinaro to OpenAI as its Head of Preparedness.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “our Head of Preparedness” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through X’s own syndication endpoint for the post’s text, which truncates long notes. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. No human editorial review.","retrievalRoute":"x_syndication"},{"id":"lg-ev-oai-scandinaro-preparedness-2026-2","claimId":"lg-oai-scandinaro-preparedness-2026","sourceId":"lg-src-oai-scandinaro-x-2026","sourceRole":"primary","support":"supports","locator":"Post text, first two sentences","passage":"Scandinaro writes that he is joining OpenAI as Head of Preparedness and thanks colleagues from his time at Anthropic.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “joining OpenAI as Head of Preparedness” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through X’s own syndication endpoint for the post’s text, which truncates long notes. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. No human editorial review.","retrievalRoute":"x_syndication"},{"id":"lg-ev-ant-mann-openai-prior","claimId":"lg-ant-mann-openai-prior","sourceId":"source-anthropic-reinvent-2024","sourceRole":"primary","support":"supports","locator":"PDF page 6, “Our business”, founder panel, Ben Mann row (row layout checked on the rendered page)","passage":"Mann's row describes him as a former GPT-3 and API engineer at OpenAI.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “Former GPT-3 and API Engineer at OpenAI” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-cn-zhang-peng-zhipu-cofounder","claimId":"lg-cn-zhang-peng-zhipu-cofounder","sourceId":"lg-src-cn-zhipu-prospectus-en","sourceRole":"primary","support":"supports","locator":"Business, description of management team, printed page 154","passage":"Dr. Zhang Peng, our co-founder, executive Director and CEO","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (9 words) matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"ev-cmp-preparedness-beta-2023-d3-safeguards-1","claimId":"cmp-preparedness-beta-2023-d3-safeguards","sourceId":"source-pf-beta","sourceRole":"primary","support":"supports","locator":"PDF p. 20, Governance › Safety baselines › 'Asset Protection', first sentence","passage":"we will ensure that our security is hardened in a way that is designed to prevent our mitigations and controls from being circumvented via exfiltration","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-27 against the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. Human editorial review is pending."},{"id":"evidence-ap-john-diebold-diebold-1955-automation-condensation","claimId":"ap-john-diebold-diebold-1955-automation-condensation","sourceId":"source-diebold-1955-automation-condensation","sourceRole":"primary","support":"supports","locator":"Printed page 14, heading of the condensation and the permission note before it, which credits the Controllership Foundation with the condensation (OCR text of the scanned volume)","passage":"A Condensation of the Book Automation, by John Diebold","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Checked by an automated research assistant on 2026-10-27 in a direct retrieval of the Internet Archive scan’s OCR text and by a reviewer from a different model family on 2026-10-27, who distinguished the book’s author from the condensation’s preparer; the page image was not inspected. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"evidence-ap-chris-van-merwijk-rlo-2019-deceptive-alignment","claimId":"ap-chris-van-merwijk-rlo-2019-deceptive-alignment","sourceId":"source-rlo-2019-deceptive-alignment","sourceRole":"primary","support":"supports","locator":"Post author list in the publisher’s post data (LessWrong GraphQL API)","passage":"Chris van Merwijk","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval through the publisher’s post data and checked by a reviewer from a different model family on 2026-10-27. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"evidence-ap-eliezer-yudkowsky-extropians-1998-09-07-yudkowsky-2209","claimId":"ap-eliezer-yudkowsky-extropians-1998-09-07-yudkowsky-2209","sourceId":"source-extropians-1998-09-07-yudkowsky-2209","sourceRole":"primary","support":"supports","locator":"Message header, sender name (address not reproduced)","passage":"Eliezer S. Yudkowsky","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"evidence-ap-eliezer-yudkowsky-sl4-2002-06-23-yudkowsky-0206-4219","claimId":"ap-eliezer-yudkowsky-sl4-2002-06-23-yudkowsky-0206-4219","sourceId":"source-sl4-2002-06-23-yudkowsky-0206-4219","sourceRole":"primary","support":"supports","locator":"Message header, sender name (address not reproduced)","passage":"Eliezer S. Yudkowsky","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval and checked by a reviewer from a different model family on 2026-10-27. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"ev-cmp-rsp-v2-1-d8-process-1","claimId":"cmp-rsp-v2-1-d8-process","sourceId":"source-rsp-2-1","sourceRole":"primary","support":"supports","locator":"PDF p. 17 (printed p. 13), §7.1 Internal Governance, numbered item 7 ('Policy changes'), first sentence","passage":"The edition says the CEO and the Responsible Scaling Officer will propose changes to the policy and the Board of Directors will approve them, consulting the Long-Term Benefit Trust.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-27","note":"Explicit paraphrase, written on 2026-10-27 by an automated research assistant after re-reading the passage in the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. The comparison quoted the passage; it is given here as a paraphrase to keep the edition within the quotation limit. The subject is changes to this policy. §5 (PDF p. 14) adds that, on an upgrade to Required Safeguards, the policy is to be updated with any further Capability Thresholds for which those safeguards would not be sufficient. Human editorial review is pending."},{"id":"c2-ev-baidu-20f-fy2023-genai-filing","claimId":"c2-baidu-20f-fy2023-genai-filing","sourceId":"c2-src-baidu-20f-fy2023","sourceRole":"primary","support":"supports","locator":"Item 4.B, Regulations, section on generative AI","passage":"We had completed the necessary security assessment and algorithm filing in accordance with the Generative AI Services Measures and other applicable laws and regulations before we launched Ernie Bot to the public.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (32 words) matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day. Read through the third-party reader relay r.jina.ai, which renders the publisher’s current page (direct retrieval failed); not an archive copy. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-27, item K004) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"reader_relay"},{"id":"c2-ev-governance-tc260-asgf-v2-loss-of-control-testing-1","claimId":"c2-governance-tc260-asgf-v2-loss-of-control-testing","sourceId":"c2-src-tc260-asgf-v2-pdf","sourceRole":"primary","support":"supports","locator":"§5.11 增进协同应对人工智能失控风险的共识","passage":"开发者定期进行测试，判断模型是否可能带来潜在技术失控风险","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). Transcribed as printed. The passage matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day, read directly from the publisher; in the PDF’s text layer it contains control characters (U+0001) and compatibility forms (Kangxi radicals, ⻛ for 风, ⻅ for 见), which the candidate pack marked with ellipses. The cross-family check found that those ellipses do not reproduce the printed text (K047), so the quotation was re-extracted at admission, without them and with the forms normalised, from a direct retrieval of the PDF whose SHA-256 equals the pack’s. English machine translation, not verified by a person: “developers regularly conduct tests to judge whether a model may bring potential risks of technical loss of control”. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-eu-aia-art52-5-reassessment-1","claimId":"fl-eu-aia-art52-5-reassessment","sourceId":"source-eu-ai-act-2024","sourceRole":"primary","support":"supports","locator":"Article 52(5), third sentence","passage":"Providers may request reassessment at the earliest six months after the designation decision.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (13 words) matched the text retrieved for the candidate pack on 2026-10-27 and again at admission on 2026-10-28, in a fresh retrieval whose SHA-256 equals the pack’s. Read the Publications Office's Cellar repository (the same Official Journal expression as the dataset's EUR-Lex source; EUR-Lex returned empty responses). Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-eu-aia-art52-5-reassessment-2","claimId":"fl-eu-aia-art52-5-reassessment","sourceId":"source-eu-ai-act-2024","sourceRole":"primary","support":"supports","locator":"Article 52(5), first, second and fourth sentences","passage":"The Commission takes a reasoned request into account and may decide to reassess; the request must contain objective, detailed and new reasons arising since the designation; after a decision to maintain the designation, a further request is possible at the earliest six months later.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The paraphrase is Selekoa’s, written from the text retrieved on 2026-10-27 and read the Publications Office's Cellar repository (the same Official Journal expression as the dataset's EUR-Lex source; EUR-Lex returned empty responses); it is not a quotation and was not machine-matched. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-uk-cohere-mou-not-binding-1","claimId":"fl-uk-cohere-mou-not-binding","sourceId":"source-uk-cohere-mou-2025","sourceRole":"primary","support":"supports","locator":"Final paragraph","passage":"This memorandum is voluntary, not legally binding, and without prejudice to any binding agreements.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (14 words) matched the text retrieved for the candidate pack on 2026-10-27 and again at admission on 2026-10-28, in a fresh retrieval of the same address (the page’s bytes had changed). Read directly from www.gov.uk. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. The 'AI Security' section says the parties 'may' explore research collaborations with the AI Security Institute (paraphrase). No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-us-eo14299-reactors-for-ai-1","claimId":"cpo-pw-us-eo14299-reactors-for-ai","sourceId":"source-cpo-pw-eo-14299-fr","sourceRole":"primary","support":"supports","locator":"Sec. 4(c), 90 FR 22582","passage":"for the purpose of powering AI infrastructure, other critical or national security needs","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (13 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 50) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-us-eo14299-reactors-for-ai-2","claimId":"cpo-pw-us-eo14299-reactors-for-ai","sourceId":"source-cpo-pw-eo-14299-fr","sourceRole":"primary","support":"supports","locator":"Sec. 4(c), 90 FR 22582","passage":"with a goal of operating an advanced nuclear reactor at the first site no later than 30 months from the date of this order","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (24 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 50) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-google-constellation-890mw-1","claimId":"cpo-pw-google-constellation-890mw","sourceId":"source-cpo-pw-constellation-google-890-2026","sourceRole":"primary","support":"supports","locator":"paragraph beginning “BALTIMORE and MOUNTAIN VIEW, Calif. —…”","passage":"bring 890 megawatts (MW) of new, reliable nuclear capacity onto the PJM Interconnection grid","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (14 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 37) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-google-constellation-890mw-2","claimId":"cpo-pw-google-constellation-890mw","sourceId":"source-cpo-pw-constellation-google-890-2026","sourceRole":"primary","support":"supports","locator":"paragraph beginning “The 20-year power purchase agreement will…”","passage":"The 20-year power purchase agreement will enable investments in new equipment and technology at 11 Constellation-owned nuclear units in Illinois, Pennsylvania, and New Jersey","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (24 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 37) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-google-constellation-890mw-3","claimId":"cpo-pw-google-constellation-890mw","sourceId":"source-cpo-pw-constellation-google-890-2026","sourceRole":"primary","support":"supports","locator":"list 'Key elements of the comprehensive commercial agreement', item 'Direct supply expansion'","passage":"with the first uprate expected to be delivered by 2028","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (10 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 37) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-google-constellation-890mw-4","claimId":"cpo-pw-google-constellation-890mw","sourceId":"source-cpo-pw-constellation-google-890-2026","sourceRole":"primary","support":"supports","locator":"paragraph beginning “The agreement offers a direct response…”","passage":"to support its growth without passing on costs to residential customers","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (11 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 37) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"pp-ev-er-giannandrea-apple-svp-2025-1","claimId":"pp-er-giannandrea-apple-svp-2025","sourceId":"lg-src-gw-apple-giannandrea-2025","sourceRole":"primary","support":"supports","locator":"First paragraph","passage":"John Giannandrea, Apple’s senior vice president for Machine Learning and AI Strategy, is stepping down from his position and will serve as an advisor to the company before retiring in the spring of 2026.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Also the dataset's departure passage. The whole sentence, already quoted from this release by the dataset’s departure record, is given in place of the pack’s shorter quotation of its opening, so that the release’s quotation counts once. The quotation (18 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes, PDF line-break hyphens and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-30, item 34) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-aib-ppl-rives-esm1b-1","claimId":"aib-ppl-rives-esm1b","sourceId":"source-aib-esm1b-paper","sourceRole":"primary","support":"supports","locator":"author list","passage":"Find articles by Alexander Rives","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI in biology (research/ai-in-biology). The quotation (5 words) matched the text retrieved for the candidate pack on 2026-10-30. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-aib-ppl-rives-esm1b-2","claimId":"aib-ppl-rives-esm1b","sourceId":"source-aib-esm1b-paper","sourceRole":"primary","support":"supports","locator":"Author contributions","passage":"A.R., J. Meier, T.S., S.G., Z.L., M.O., C.L.Z., J. Ma, and R.F. designed research","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI in biology (research/ai-in-biology). The quotation (14 words) matched the text retrieved for the candidate pack on 2026-10-30. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-kp-states-mmlu-pro-more-robust-challenging-2024-1-1","claimId":"kp-states-mmlu-pro-more-robust-challenging-2024-1","sourceId":"source-kp-mmlu-pro-more-robust-challenging-2024-v1","sourceRole":"primary","support":"supports","locator":"arXiv v1, PDF p. 1, abstract","passage":"This paper introduces MMLU-Pro, an enhanced dataset designed to extend the mostly knowledge-driven MMLU benchmark by integrating more challenging, reasoning-focused questions and expanding the choice set from four to ten options.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The quotation (31 words) matched the text retrieved for the candidate pack on 2026-10-28 (research/key-papers/verification.json). Matching ignored differences of whitespace, case, quotation marks, dashes and line-break hyphens, so the source’s typography can differ. A cross-family machine check (2026-10-30, item B047) found the quotation at its locator. No human editorial review."},{"id":"evidence-kp-states-adversarial-winograd-schema-challenge-at-2019-1-1","claimId":"kp-states-adversarial-winograd-schema-challenge-at-2019-1","sourceId":"source-kp-adversarial-winograd-schema-challenge-at-2019-v1","sourceRole":"primary","support":"supports","locator":"arXiv v1, PDF p. 1, abstract","passage":"We introduce W INO G RANDE, a new collection of WSC problems that are adversarially constructed to be robust against spurious statistical biases.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The quotation (23 words) matched the text retrieved for the candidate pack on 2026-10-28 (research/key-papers/verification.json). Matching ignored differences of whitespace, case, quotation marks, dashes and line-break hyphens, so the source’s typography can differ. A cross-family machine check (2026-10-30, item B050) found the quotation at its locator. No human editorial review."},{"id":"evidence-kp-cites-iem-2013-ai-positive-negative-factor-global-risk-1","claimId":"kp-cites-iem-2013-ai-positive-negative-factor-global-risk","sourceId":"source-iem","sourceRole":"primary","support":"supports","locator":"PDF p. 95, reference or mention by title","passage":"The text names the work by its title.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The citation was matched (title) in the text retrieved for the candidate pack on 2026-10-28 and re-found by its mechanical check. The paraphrase is Selekoa’s. No human editorial review."},{"id":"evidence-kp-cites-ernie-5-report-deepseek-r1-incentivizing-reasoning-capability-2025-1","claimId":"kp-cites-ernie-5-report-deepseek-r1-incentivizing-reasoning-capability-2025","sourceId":"c2-src-ernie5-arxiv-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 31, reference by arXiv identifier","passage":"The text gives the arXiv identifier 2501.12948.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The citation was matched (arXiv identifier) in the text retrieved for the candidate pack on 2026-10-28 and re-found by its mechanical check. The paraphrase is Selekoa’s. No human editorial review."},{"id":"evidence-kp-cites-acemoglu-restrepo-2019-wrong-kind-of-ai-robots-jobs-evidence-from-us-2019-1","claimId":"kp-cites-acemoglu-restrepo-2019-wrong-kind-of-ai-robots-jobs-evidence-from-us-2019","sourceId":"source-acemoglu-restrepo-2019-wrong-kind-of-ai","sourceRole":"primary","support":"supports","locator":"PDF p. 13, reference or mention by title","passage":"The text names the work by its title.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The citation was matched (title) in the text retrieved for the candidate pack on 2026-10-28 and re-found by its mechanical check. The paraphrase is Selekoa’s. No human editorial review."},{"id":"evidence-kp-cites-source-gemini-report-v1-sequence-sequence-learning-neural-networks-2014-1","claimId":"kp-cites-source-gemini-report-v1-sequence-sequence-learning-neural-networks-2014","sourceId":"source-gemini-report-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 32, reference or mention by title","passage":"The text names the work by its title.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The citation was matched (title) in the text retrieved for the candidate pack on 2026-10-28 and re-found by its mechanical check. The paraphrase is Selekoa’s. No human editorial review."},{"id":"evidence-kp-authored-laurent-sifre-mastering-game-go-deep-neural-2016-1","claimId":"kp-authored-laurent-sifre-mastering-game-go-deep-neural-2016","sourceId":"source-kp-mastering-game-go-deep-neural-2016-text","sourceRole":"primary","support":"supports","locator":"page head, byline","passage":"Laurent Sifre","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The printed name (2 words) was found in the byline of the text retrieved for the candidate pack on 2026-10-28. No human editorial review."},{"id":"evidence-kp-authored-dario-amodei-amplification-paper-1","claimId":"kp-authored-dario-amodei-amplification-paper","sourceId":"source-amplification-paper-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 1, byline","passage":"Dario Amodei","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The printed name (2 words) was found in the byline of the text retrieved for the candidate pack on 2026-10-28. No human editorial review."},{"id":"evidence-th-term-fm-bommasani-2021-naming-1","claimId":"th-term-fm-bommasani-2021-naming","sourceId":"source-kp-opportunities-risks-foundation-models-2021-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 6, section 1.1.1 \"Naming\"","passage":"Existing terms (e.g., pretrained model, self-supervised model) partially capture the technical dimension of these models","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, Kinds of AI and technical history (research/technical-history). The quotation (15 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes, PDF line-break hyphens and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-th-term-fm-bommasani-2021-naming-2","claimId":"th-term-fm-bommasani-2021-naming","sourceId":"source-kp-opportunities-risks-foundation-models-2021-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 7, section 1.1.1 \"Naming\"","passage":"Having surveyed existing terms and numerous alternatives, we chose to coin the new term foundation models to identify the models, and the emerging paradigm, that are the subject of this report.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, Kinds of AI and technical history (research/technical-history). The quotation (31 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes, PDF line-break hyphens and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"c3-ev-iflytek-liu-qingfeng-president-2008","claimId":"c3-iflytek-liu-qingfeng-president-2008","sourceId":"c3-src-iflytek-prospectus-2008","sourceRole":"primary","support":"supports","locator":"Section 8, Directors, supervisors and senior management, (1) Directors","passage":"刘庆峰先生，董事、总裁，中国国籍","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, Chinese frontier-AI institutions, third collection (research/china-labs-3). The quotation (13 words, each CJK character counted as one) matched the text retrieved for the candidate pack on 2026-10-29 and again in a fresh retrieval the same day, read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. English machine translation, not verified by a person: “Mr. Liu Qingfeng, director and president, Chinese nationality”. A cross-family machine check (2026-10-31, item K004) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r06-leike-2022-optimistic-alignment-approach-1","claimId":"rd-r06-leike-2022-optimistic-alignment-approach","sourceId":"source-rd-leike-2022-optimistic-alignment-approach","sourceRole":"primary","support":"supports","locator":"Summary of the five sources","passage":"A lot of developments over the last few years have made AI systems more favorable to alignment than they looked initially","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (21 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r06-leike-2022-optimistic-alignment-approach-2","claimId":"rd-r06-leike-2022-optimistic-alignment-approach","sourceId":"source-rd-leike-2022-optimistic-alignment-approach","sourceRole":"primary","support":"supports","locator":"Summary of the five sources","passage":"We're just trying to align a system that's capable enough to make more alignment progress than we can.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (18 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r06-leike-2022-optimistic-alignment-approach-3","claimId":"rd-r06-leike-2022-optimistic-alignment-approach","sourceId":"source-rd-leike-2022-optimistic-alignment-approach","sourceRole":"primary","support":"context","locator":"Section 2, \"A more modest goal\"","passage":"Finding this once-and-for-all solution could be very difficult, and we don't really know how to do this.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (17 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r07-hawking-russell-tegmark-wilczek-2014-independent-1","claimId":"rd-r07-hawking-russell-tegmark-wilczek-2014-independent","sourceId":"source-rd-hawking-russell-tegmark-wilczek-2014-independent","sourceRole":"primary","support":"supports","locator":"Paragraphs 3–4","passage":"Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last, unless we learn how to avoid the risks","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (27 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r07-hawking-russell-tegmark-wilczek-2014-independent-2","claimId":"rd-r07-hawking-russell-tegmark-wilczek-2014-independent","sourceId":"source-rd-hawking-russell-tegmark-wilczek-2014-independent","sourceRole":"primary","support":"supports","locator":"Paragraph 6","passage":"the short-term impact of AI depends on who controls it, the long-term impact depends on whether it can be controlled at all.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (22 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r12-nature-2023-editorial-stop-talking-1","claimId":"rd-r12-nature-2023-editorial-stop-talking","sourceId":"source-rd-nature-2023-editorial-stop-talking","sourceRole":"primary","support":"supports","locator":"Standfirst","passage":"Talk of artificial intelligence destroying humanity plays into the tech companies’ agenda, and hinders effective regulation of the societal harms AI is causing right now.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (25 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r12-nature-2023-editorial-stop-talking-2","claimId":"rd-r12-nature-2023-editorial-stop-talking","sourceId":"source-rd-nature-2023-editorial-stop-talking","sourceRole":"primary","support":"supports","locator":"Paragraph 3","passage":"it draws attention away from the real issue: the societal harms that AI systems and tools are causing now, or risk causing in future.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (24 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r14-bender-koller-2020-climbing-nlu-1","claimId":"rd-r14-bender-koller-2020-climbing-nlu","sourceId":"source-rd-bender-koller-2020-climbing-nlu","sourceRole":"primary","support":"supports","locator":"Abstract, PDF p. 1","passage":"we argue that a system trained only on form has a priori no way to learn meaning","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (17 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-r14-bender-koller-2020-climbing-nlu-2","claimId":"rd-r14-bender-koller-2020-climbing-nlu","sourceId":"source-rd-bender-koller-2020-climbing-nlu","sourceRole":"primary","support":"supports","locator":"Section 1, PDF p. 1","passage":"the language modeling task, because it only uses form as training data, cannot in principle lead to learning of meaning","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (20 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"}],"at":{"entities":[0,1,3,15,16,19,21,23,33,39,42,43,44,45,46,47,48,49,53,54,81,99,126,176,187,191,207,214,250,251,252,267,301,314,331,346,357,362,445,470,472,491,512,513,522,533,546,548,569,579,631,635,638,663,707,724,758,872,932,938,952,959,1002,1003,1004,1032,1074,1140,1310,1341,1444,1456,1473,1476,1479,1495,1531,1670,1773,1776,1789,1842,1863,1864,1888,1900],"claims":[43,88,178,363,428,661,692,714,802,1045,1046,1076,1290,1303,1465,1695,1844,1932,1955,2120,2218,2342,2358,2411,2515,2739,2810,2870,3033,3050,3178,3476,3651,3795,3803,4022,4069,4258,4459,4638,4676,4758,4915,5191,5195,5260,5277],"sources":[10,17,23,37,43,48,70,81,98,121,163,212,222,225,235,239,267,304,336,412,437,439,457,484,498,500,522,532,587,670,757,758,763,774,786,832,867,934,1007,1189,1273,1312,1327,1364,1575,1708,1750,1766,1787],"evidence":[45,105,197,386,460,820,821,869,893,1012,1273,1274,1311,1617,1632,1828,2141,2142,2143,2372,2373,2496,2528,2749,2865,2989,3005,3058,3163,3447,3580,3679,3680,3938,3976,3977,4260,4261,4262,4263,4793,5044,5045,5253,5261,5480,5527,5716,5917,6096,6134,6225,6226,6443,6914,6915,6916,6925,6926,7034,7035,7063,7064]},"version":"131evrz"}