{"index":{"byClaim":[["prov-fsf-v2-0-hold-pending-mitigations",["documents:frontier-safety-frameworks"]],["absent-rsp-v3-4-no-train-deploy-without-safeguards",["documents:frontier-safety-frameworks"]],["sb-attorney-general-enforcement",["commitments:external-commitments","guide:frontier-frameworks-and-law"]],["cs-openai-group-pbc-legal-form-2025",["control:openai"]],["ts-a3-iem",["propositions:superintelligence"]],["ts-b3-agi-ruin",["propositions:superintelligence"]],["ec-g06-clancy-besiroglu-2023-great-inflection-a",["propositions:economics-of-machine-intelligence"]],["ec-m05-acemoglu-restrepo-2016-race-between-machine-and-man-a",["propositions:economics-of-machine-intelligence"]],["ec-m06b-agrawal-gans-goldfarb-2018-economic-policy-for-ai-a",["propositions:economics-of-machine-intelligence"]],["ec-m09-nordhaus-2015-economic-singularity-b",["propositions:economics-of-machine-intelligence"]],["ec-m09-calvano-et-al-2020-algorithmic-pricing-collusion-b",["propositions:economics-of-machine-intelligence"]],["lg-ant-cuellar-ltbt-trustee",["control:anthropic"]],["cmp-fsf-v2-0-d6-discretion",["compare:frontier-safety-frameworks"]],["cmp-rsp-v2-0-d5-cadence",["compare:frontier-safety-frameworks"]],["cmp-rsp-v2-1-d1-other",["compare:frontier-safety-frameworks"]],["fl-eu-aia-annex13-criteria",["guide:frontier-frameworks-and-law"]]],"marks":[["clancy-besiroglu-2023-great-inflection",[{"collectionId":"economics-of-machine-intelligence","propositionId":"explosive-growth-from-labour-substitution","claimId":"ec-g01-clancy-besiroglu-2023-great-inflection-c","text":"Earliest located tier-2 text contesting G01 Machine substitution for labour produces explosive growth in Economics of machine intelligence"},{"collectionId":"economics-of-machine-intelligence","propositionId":"unautomated-tasks-bottleneck-growth","claimId":"ec-g04-clancy-besiroglu-2023-great-inflection-a","text":"Earliest located in tier 2 of Economics of machine intelligence for G04 Essential unautomated tasks bottleneck growth"}]],["acemoglu-restrepo-2016-race-between-machine-and-man",[{"collectionId":"economics-of-machine-intelligence","propositionId":"machines-cheaper-at-every-task","claimId":"ec-m05-acemoglu-restrepo-2016-race-between-machine-and-man-a","text":"Earliest located in tier 1 of Economics of machine intelligence for M05 Machines cheaper than labour at every task"}]],["nordhaus-2015-economic-singularity",[{"collectionId":"economics-of-machine-intelligence","propositionId":"unautomated-tasks-bottleneck-growth","claimId":"ec-g04-nordhaus-2015-economic-singularity-a","text":"Earliest located in tier 1 of Economics of machine intelligence for G04 Essential unautomated tasks bottleneck growth"},{"collectionId":"economics-of-machine-intelligence","propositionId":"other-constraints-prevent-explosive-growth","claimId":"ec-g05-nordhaus-2015-economic-singularity-c","text":"Earliest located tier-1 text contesting G05 Other constraints prevent explosive growth in Economics of machine intelligence"},{"collectionId":"economics-of-machine-intelligence","propositionId":"ai-existential-risk-in-economic-analysis","claimId":"ec-e03-nordhaus-2015-economic-singularity-a","text":"Earliest located in tier 1 of Economics of machine intelligence for E03 Existential risk from AI as a matter for economic analysis"},{"collectionId":"economics-of-machine-intelligence","propositionId":"economics-engagement-with-superintelligence","claimId":"ec-e05-nordhaus-2015-economic-singularity-a","text":"Earliest located in tier 1 of Economics of machine intelligence for E05 Economics’ engagement with superintelligence, AGI and the singularity"}]],["kp-basic-ai-drives-2008",[{"collectionId":"risk-debate","propositionId":"benevolence-or-no-drive-to-dominate","claimId":"rd-r10a-omohundro-2008-basic-ai-drives","text":"Earliest located text contesting R10a Intelligence brings benevolence or no drive to dominate in Arguments about catastrophic risk from advanced AI"}]]],"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":[]},"collections":[{"target":{"kind":"propositions","id":"superintelligence"},"title":"Propositions about superintelligence, 1863–2022","claimIds":["ts-a3-iem","ts-b3-agi-ruin"],"records":[]},{"target":{"kind":"propositions","id":"risk-debate"},"title":"Arguments about catastrophic risk from advanced AI","claimIds":[],"records":[]},{"target":{"kind":"propositions","id":"yudkowsky-arguments"},"title":"Eliezer Yudkowsky’s arguments on common claims about AI, 1996–2026","claimIds":[],"records":[]},{"target":{"kind":"propositions","id":"economics-of-machine-intelligence"},"title":"Economics of machine intelligence","claimIds":["ec-g06-clancy-besiroglu-2023-great-inflection-a","ec-m05-acemoglu-restrepo-2016-race-between-machine-and-man-a","ec-m06b-agrawal-gans-goldfarb-2018-economic-policy-for-ai-a","ec-m09-nordhaus-2015-economic-singularity-b","ec-m09-calvano-et-al-2020-algorithmic-pricing-collusion-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":["prov-fsf-v2-0-hold-pending-mitigations","absent-rsp-v3-4-no-train-deploy-without-safeguards"]},{"target":{"kind":"compare","id":"frontier-safety-frameworks"},"title":"Compare frameworks","claimIds":["cmp-fsf-v2-0-d6-discretion","cmp-rsp-v2-0-d5-cadence","cmp-rsp-v2-1-d1-other"]},{"target":{"kind":"control","id":"openai"},"title":"OpenAI’s control structure, 2015–2026","claimIds":["cs-openai-group-pbc-legal-form-2025"]},{"target":{"kind":"control","id":"anthropic"},"title":"Anthropic’s control structure, 2021–2026","claimIds":["lg-ant-cuellar-ltbt-trustee"]},{"target":{"kind":"commitments"},"title":"External commitments and obligations","claimIds":["sb-attorney-general-enforcement"]},{"target":{"kind":"question","id":"frameworks-to-obligations"},"title":"Frameworks, commitments and legal obligations, 2023–2026","claimIds":["sb-attorney-general-enforcement"]},{"target":{"kind":"reading"},"title":"Founding and early leadership","claimIds":[]}]},"entities":[{"id":"openai","name":"OpenAI","type":"organisation"},{"id":"xai","name":"xAI","type":"organisation"},{"id":"nvidia","name":"NVIDIA","type":"organisation"},{"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":"fsf-2024","type":"document","documentVersion":{"family":"deepmind-fsf","label":"v1.0","publisherId":"google-deepmind","publishedAt":"2024-05-17","dateNote":"The sourced publication date does not establish an implementation or effective date."}},{"id":"iem-2013","name":"Intelligence Explosion Microeconomics","type":"document"},{"id":"deceptive-alignment","name":"Deceptive alignment","type":"concept","proposition":{"code":"B3","criterion":"A trained or learned system that is not aligned behaves as if aligned during training because doing so serves another objective, for example to avoid modification and act on that objective later.","terms":["deceptive alignment"]}},{"id":"google","name":"Google","type":"organisation"},{"id":"rsp-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."},"name":"Anthropic Responsible Scaling Policy — v2.0"},{"id":"rsp-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."},"name":"Anthropic Responsible Scaling Policy — v2.1"},{"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","name":"Anthropic Responsible Scaling Policy — 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."}},{"id":"fsf-v2-0","name":"Google DeepMind Frontier Safety Framework — v2.0","type":"document","documentVersion":{"family":"deepmind-fsf","label":"v2.0","publisherId":"google-deepmind","publishedAt":"2025-02-04","dateNote":"The sourced publication date does not establish an implementation or effective date."}},{"id":"fsf-v3-0","type":"document","documentVersion":{"family":"deepmind-fsf","label":"v3.0","publisherId":"google-deepmind","publishedAt":"2025-09-22","dateNote":"The sourced publication date does not establish an implementation or effective date."}},{"id":"fsf-v3-1","type":"document","documentVersion":{"family":"deepmind-fsf","label":"v3.1","publisherId":"google-deepmind","publishedAt":"2026-04-17","dateNote":"The sourced publication date does not establish an implementation or effective date."}},{"id":"andrej-karpathy","name":"Andrej Karpathy","type":"person"},{"id":"attention-all-you-need","name":"Attention Is All You Need","type":"document"},{"id":"ppo-paper","name":"Proximal Policy Optimization Algorithms","type":"document"},{"id":"self-critique-paper","name":"Self-critiquing models for assisting human evaluators","type":"document"},{"id":"spacex","name":"SpaceX","type":"organisation"},{"id":"llama-paper","name":"LLaMA (2023 paper)","type":"document"},{"id":"qwen-team","name":"Qwen Team / 通义千问","type":"organisation"},{"id":"junyang-lin","name":"Junyang Lin","type":"person"},{"id":"chatglm-report","name":"ChatGLM research history and techniques","type":"document"},{"id":"california-sb-53","name":"California SB 53 (2025): Transparency in Frontier Artificial Intelligence Act, as chaptered","type":"document","documentVersion":{"family":"ca-sb-53","label":"Chaptered (Stats. 2025, ch. 138)","publisherId":"california-legislature","publishedAt":"2025-09-29","effectiveAt":"2026-01-01","dateNote":"Approved by the Governor and filed with the Secretary of State on 2025-09-29 (chaptered header and bill history); the chaptered text page reads “Date Published: 09/29/2025 09:00 PM”. The added code sections took effect on 2026-01-01 according to the codified history notes checked on 2026-10-04, which showed no amendment. Gov. Code § 11546.8 is operative only on an appropriation."}},{"id":"openai-group-pbc","name":"OpenAI Group PBC","type":"organisation"},{"id":"anthropic-ltbt","name":"Anthropic Long-Term Benefit Trust","type":"organisation"},{"id":"agi-ruin-2022","name":"AGI Ruin: A List of Lethalities","type":"document"},{"id":"recursive-self-improvement","name":"Recursive self-improvement","type":"concept","proposition":{"code":"A3","criterion":"A system improves its own capacity to improve itself, so that each improvement feeds the next (self-augmentation, self-enhancement). Learning from experience or teaching alone does not meet the criterion, and neither does building a better successor, or adding machines to a population, unless the same system is improved.","terms":["recursive self-improvement"]}},{"id":"lw-2022-six-dimensions-of-operational-adequacy-in","name":"Six Dimensions of Operational Adequacy in AGI Projects","type":"document"},{"id":"sl4-2000-08-21-yudkowsky-0008-0046","name":"Re: [SL4] Employment vs. 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":"no-current-sign-of-explosive-growth","name":"Current data show no approach to explosive growth","type":"concept","proposition":{"code":"G06","criterion":"The text states, on the basis of current or recent economic data, that the economy shows no sign yet of approaching a growth singularity or explosive growth.","terms":[]}},{"id":"machines-cheaper-at-every-task","name":"Machines cheaper than labour at every task","type":"concept","proposition":{"code":"M05","criterion":"A text that states or models the possibility that machines perform all or nearly all tasks, including tasks created later, at lower cost than human labour. Texts that assume new tasks in which labour is cheaper keep appearing bear on it without meeting it.","terms":[]}},{"id":"compensating-policy-for-ai","name":"Compensating policy for AI’s distributional effects","type":"concept","proposition":{"code":"M06b","criterion":"A text that argues for redistribution, taxation or another compensating policy in response to distributional effects of AI or automation.","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":"nordhaus-2015-economic-singularity","name":"Are We Approaching an Economic Singularity? Information Technology and the Future of Economic Growth","type":"document"},{"id":"clancy-besiroglu-2023-great-inflection","name":"The Great Inflection? A Debate About AI and Explosive Growth","type":"document"},{"id":"acemoglu-restrepo-2016-race-between-machine-and-man","name":"The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment","type":"document"},{"id":"agrawal-gans-goldfarb-2018-economic-policy-for-ai","name":"Economic Policy for Artificial Intelligence","type":"document"},{"id":"calvano-et-al-2020-algorithmic-pricing-collusion","name":"Artificial Intelligence, Algorithmic Pricing, and Collusion","type":"document"},{"id":"arvind-kc","name":"Arvind KC","type":"person"},{"id":"laurent-sifre","name":"Laurent Sifre","type":"person"},{"id":"mariano-florentino-cuellar","name":"Mariano-Florentino Cuéllar","type":"person"},{"id":"niki-parmar","name":"Niki Parmar","type":"person"},{"id":"ross-nordeen","name":"Ross Nordeen","type":"person"},{"id":"sensetime","name":"SenseTime / 商汤科技","type":"organisation"},{"id":"stepfun","name":"StepFun / 阶跃星辰","type":"organisation"},{"id":"windsurf","name":"Windsurf","type":"organisation"},{"id":"kimi-k2-report","name":"Kimi K2: Open Agentic Intelligence","type":"document"},{"id":"spacex-orbital-data-center-system","name":"SpaceX Orbital Data Center system (proposed)","type":"infrastructure"},{"id":"kp-basic-ai-drives-2008","name":"The Basic AI Drives","type":"document"},{"id":"kp-gpt-4-technical-report-2023","name":"GPT-4 Technical Report","type":"document"},{"id":"kp-constitutional-ai-harmlessness-from-ai-2022","name":"Constitutional AI: Harmlessness from AI Feedback","type":"document"},{"id":"kp-triviaqa-large-scale-distantly-supervised-2017","name":"TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension","type":"document"},{"id":"kp-scaling-language-models-methods-analysis-2021","name":"Scaling Language Models: Methods, Analysis & Insights from Training Gopher","type":"document"},{"id":"kp-self-consistency-improves-chain-thought-2022","name":"Self-Consistency Improves Chain of Thought Reasoning in Language Models","type":"document"},{"id":"kp-xgboost-scalable-tree-boosting-system-2016","name":"XGBoost: A Scalable Tree Boosting System","type":"document"},{"id":"alec-radford","name":"Alec Radford","type":"person"},{"id":"th-zhao-2023-llm-survey","name":"A Survey of Large Language Models","type":"document"},{"id":"th-term-large-language-model","name":"'large language model'","type":"concept"},{"id":"xu-li","name":"Xu Li","type":"person"},{"id":"step1x-edit-report","name":"Step1X-Edit: A Practical Framework for General Image Editing","type":"document"},{"id":"cn-cac-genai-service-list-2026-08","name":"生成式人工智能服务已备案和已登记信息（2026年8月） — CAC list of generative AI services, attachment 13","type":"document"},{"id":"aschenbrenner-2024-situational-awareness","name":"Situational Awareness: The Decade Ahead","type":"document"},{"id":"narayanan-kapoor-2024-ai-scaling-myths","name":"AI scaling myths","type":"document"}],"claims":[{"id":"prov-fsf-v2-0-hold-pending-mitigations","subject":"fsf-v2-0","predicate":"states_provision","object":{"kind":"literal","value":"Response plan may put deployment or development on hold"},"topic":"governance","statement":"FSF v2.0 states that if a model flagged by an alert threshold is assessed to pose risks for which readily available mitigations may not be sufficient, the response plan may involve putting deployment or further development on hold until adequate mitigations can be applied.","time":{"kind":"event","date":"2025-02-04","basis":"publication","note":"Provision as stated in this edition, dated by the edition’s publication 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-fsf-v2-0-hold-pending-mitigations"],"attributes":{"provision":{"kind":"required_action","key":"fsf-hold-pending-mitigations"}}},{"id":"absent-rsp-v3-4-no-train-deploy-without-safeguards","subject":"rsp-v3-4","predicate":"provision_not_found","object":{"kind":"literal","value":"Commitment not to train or deploy without adequate safeguards"},"topic":"governance","statement":"No passage stating the provision “Commitment not to train or deploy without adequate safeguards” was found in RSP · v3.4 within the inspected scope.","time":{"kind":"event","date":"2026-07-08","basis":"effective","note":"Absence recorded for this edition, dated by its effective date."},"derivation":"inferred","resolution":"supported","inferenceRationale":"Premises: RSP · v2.2 states this provision. The inspected scope of RSP · v3.4 was: RSP v3.4 (effective July 8, 2026), whole text layer, PDF pp. 1–21: Introduction (pp. 3–4); §1 with the threshold table (pp. 4–10, columns on p. 9 read individually because the text layer interleaves them); §2–§3 Risk Reports (pp. 11–16); §4 Governance (p. 16); Appendix A and Appendix B (p. 17); Changelog (pp. 18–21). Method: Same reading, comparison and case-insensitive search terms as for v3.0, plus a word-level comparison with v3.3 (differences: Automated R&D threshold, Risk Report coverage date, internal sharing, redaction disclosure, external review split). No passage stating a commitment not to train or deploy models unless adequate safeguards are implemented, or a commitment to pause training or deployment until such safeguards are in place, was found within the operative text. Hits examined and not counted: the Introduction (p. 3), §1 (p. 4) and Appendix A scenario commitments (p. 17) as in v3.0; the Appendix A introduction sentences on cautionary action and considering pausing (p. 17), as in v3.1; the reproduced RSP v2.0 changelog entry (p. 18). The conclusion is bounded by that scope and method.","editorialState":"seed","qualification":"A recorded absence within the stated scope: not a claim that the provision was deleted, and not a statement about practice or intent.","evidenceIds":["evidence-absent-rsp-v3-4-no-train-deploy-without-safeguards"],"attributes":{"provision":{"kind":"required_action","key":"rsp-no-train-deploy-without-safeguards"}}},{"id":"sb-attorney-general-enforcement","subject":"california-sb-53","predicate":"states_provision","object":{"kind":"literal","value":"Civil penalties under § 22757.15 recoverable only in a civil action brought by the Attorney General"},"topic":"governance","statement":"SB 53 provides that a civil penalty described in Bus. & Prof. Code § 22757.15 can be recovered only in a civil action brought by the California Attorney General.","time":{"kind":"event","date":"2025-09-29","basis":"publication","note":"Provision as stated in the chaptered text published and approved on 2025-09-29; the added sections took effect on 2026-01-01. Text as chaptered on 2025-09-29; the Governor approved it and it was filed with the Secretary of State that day. Effective 2026-01-01 according to the history note of codified Bus. & Prof. Code § 22757.15, which listed no amendment when checked on 2026-10-04."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the enacted text states, quoted from the chaptered expression. It is not evidence that the statute applies to any named developer, or of compliance, enforcement or effect. The Attorney General-only rule is limited to penalties described in § 22757.15. On the grounds listed in § 22757.15(a), that section makes a large frontier developer subject to a civil penalty whose amount depends on the severity of the violation and does not exceed $1,000,000 per violation. The party role 'decides' records only that the Attorney General alone may bring the civil action through which such a penalty is recovered. It does not mean that the Attorney General imposes or sets the penalty. Lab. Code § 1107.1(f)–(i), added by the same act, separately refers to civil actions and administrative proceedings 'brought pursuant to this section'. It authorises attorney's fees for a plaintiff who brings a successful action (f), sets a burden-of-proof rule (g), lets a covered employee petition the superior court for temporary or preliminary injunctive relief (h), and bars a stay of that relief pending appeal (i). Who may bring such actions is not resolved here (see stops). No enforcement action is recorded.","evidenceIds":["evidence-sb-attorney-general-enforcement-1","evidence-sb-attorney-general-enforcement-2"],"attributes":{"provision":{"kind":"decision_authority","key":"sb53-attorney-general-enforcement","parties":[{"role":"decides","actor":"Attorney General"}]}}},{"id":"cs-openai-group-pbc-legal-form-2025","subject":"openai-group-pbc","predicate":"has_legal_form","object":{"kind":"literal","value":"Delaware public benefit corporation"},"topic":"governance","statement":"OpenAI Group PBC was incorporated as a Delaware public benefit corporation. Its certificate of incorporation carries the Delaware Division of Corporations filing stamp of 28 October 2025, 08:07 AM (file number 10381551).","time":{"kind":"snapshot","date":"2025-10-28","note":"Filing date and time from the stamp on page 1 of the certificate."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Read from the filed certificate as hosted by DocumentCloud, which is byte-identical to the Internet Archive capture of 2025-10-29. The certificate states the public benefit as ensuring that artificial general intelligence benefits all of humanity (Article III). How the for-profit business came to sit in the corporation (merger, conversion or contribution) is not stated in the certificate and is not recorded.","evidenceIds":["evidence-cs-openai-group-pbc-legal-form-2025-1","evidence-cs-openai-group-pbc-legal-form-2025-2"]},{"id":"ts-a3-iem","subject":"iem-2013","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"recursive-self-improvement"},"topic":"ideas","statement":"Intelligence Explosion Microeconomics analyses the returns on reinvesting cognitive gains in better mind designs.","time":{"kind":"event","date":"2013-09-13","basis":"publication","note":"The report states \"Last modified September 13, 2013\"."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The fullest analysis of recursive improvement located; an editorial judgement.","evidenceIds":["evidence-ts-a3-iem"],"attributes":{"thought":{"roles":["most-developed"],"register":"nonfiction"}}},{"id":"ts-b3-agi-ruin","subject":"agi-ruin-2022","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"deceptive-alignment"},"topic":"ideas","statement":"AGI Ruin’s item 13 gives, as a problem that will appear at the superintelligent level, the internal behaviour of changing outer behaviour to deliberately look more aligned and deceive the programmers, operators and loss functions optimising over the system.","time":{"kind":"event","date":"2022-06-05","basis":"publication","note":"Posting time 2022-06-05T22:05:52Z; the quotes also match a capture taken about 53 minutes after posting."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An example of alignment problems that do not appear at passively safe levels of capability; item 31 treats strategic deception of people (B2).","evidenceIds":["evidence-ts-b3-agi-ruin"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction"}}},{"id":"ec-g06-clancy-besiroglu-2023-great-inflection-a","subject":"clancy-besiroglu-2023-great-inflection","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"no-current-sign-of-explosive-growth"},"topic":"ideas","statement":"Clancy and Besiroglu, “The Great Inflection? A Debate About AI and Explosive Growth” (2023-06) articulates G06. Speaker: Matt Clancy. Clancy answers Besiroglu's point about fast growth in AI training compute.","time":{"kind":"event","date":"2023-06","basis":"publication","note":"Page: \"Published June 2023\". The introduction states the conversation \"is based on a series of back-and-forths Matt and Tamay had in spring of 2023\"."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The quotation is a turn by Matt Clancy in a dialogue. An observation, not a formal test. He adds that AI is \"pretty likely to boost growth\".","evidenceIds":["evidence-ec-g06-d"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","claimed":{"date":"2023","note":"\"back-and-forths ... in spring of 2023\"; the original exchange was not retrieved and the published dialogue may be edited."},"attested":{"date":"2023-12-23","note":"Only retrieved capture."},"tier":{"value":2,"basis":"Tier 2 (unchanged). Magazine debate (Asterisk). The magazine introduces Matt Clancy as an economist and Tamay Besiroglu as a research scientist. Judgement recorded: no turn is itself a substantive scholarly argument; each answers the previous turn and argues by verbal sketches of others' models, with no model or derivation of its own, so the debate rule applies to both speakers."}}}},{"id":"ec-m05-acemoglu-restrepo-2016-race-between-machine-and-man-a","subject":"acemoglu-restrepo-2016-race-between-machine-and-man","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"machines-cheaper-at-every-task"},"topic":"ideas","statement":"Acemoglu and Restrepo, “The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment” (2016-05) articulates M05. States the full-automation case of the model, in which automation keeps pace with task creation so that labour becomes redundant; the authors immediately contrast it with the stable balanced-growth case on which they focus.","time":{"kind":"event","date":"2016-05","basis":"publication","note":"Cover \"May 2016, Revised June 2017\". Journal version: AER 108(6), 2018, pp. 1488–1542 (abstract page aer2018-ar). 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 22252, May 2016, revised June 2017; quoted wording attested from the June 2017 revision, not necessarily the May 2016 original. Published in revised form as AER 2018. Conditional model result, not a forecast; the authors treat the interior balanced-growth path as the case of interest. Holds M05's tier-1 earliest-located role since 2026-10-12, by the date of its checked wording (the June 2017 revision), not the May 2016 working-paper date: the May 2016 wording was not checked. It holds the role only as a model scenario (under the maintainer decision of 2026-10-11) meeting M05's all-tasks criterion as a conditional model; no tier-1 text in this pack states the possibility in its own voice. Simon (1965, a lead in the research pack) may, but was not retrieved. A model result under stated conditions, not the author's own claim that machines will do every task (maintainer decision of 2026-10-11). Context-only.","evidenceIds":["evidence-ec-m05-20"],"attributes":{"thought":{"roles":["earliest-located"],"register":"nonfiction","qualifier":"A model scenario under stated conditions, not the author’s own claim.","attested":{"date":"2017-06","note":"Revision date on the cover of the retrieved PDF; the May 2016 wording was not retrieved."},"tier":{"value":1,"basis":"NBER working paper (core venue)."},"conditions":"Full model with directed research toward automation versus new tasks; holds when the long-run rental rate of capital is very low relative to labour (in the formal results, a sufficiently low discount rate), giving a balanced growth path in which all tasks are produced with capital. Otherwise an interior, stable balanced growth path."}}},{"id":"ec-m06b-agrawal-gans-goldfarb-2018-economic-policy-for-ai-a","subject":"agrawal-gans-goldfarb-2018-economic-policy-for-ai","predicate":"articulates_proposition","object":{"kind":"entity","entityId":"compensating-policy-for-ai"},"topic":"ideas","statement":"Agrawal, Gans, Goldfarb, “Economic Policy for Artificial Intelligence” (2018-06) articulates M06b. Concludes that AI's consequences for jobs, inequality and competition call for education policy, the safety net and antitrust.","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 body reviews capital taxation, UBI and safety-net reinforcement, mostly via cited authors.","evidenceIds":["evidence-ec-m06b-18"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","tier":{"value":1,"basis":"NBER working paper (core venue)."}}}},{"id":"ec-m09-nordhaus-2015-economic-singularity-b","subject":"nordhaus-2015-economic-singularity","predicate":"bears_on_proposition","object":{"kind":"entity","entityId":"recent-economics-engagement-with-ai"},"topic":"ideas","statement":"Nordhaus, “Are We Approaching an Economic Singularity? Information Technology and the Future of Economic Growth” (2015-09) bears on M09. Nordhaus notes the scarcity of macroeconomic work on accelerating growth from computation and AI, apart from endogenous-growth curiosities and Simon's earlier analysis. Introducing computer scientists' views of AI and superintelligence, Nordhaus signals that these ideas lie outside economists' usual frame. Nordhaus credits Simon's 1965 book with a macroeconomic analysis of computerisation based on the factor-price frontier, and criticises its assumption of a constant capital share.","time":{"kind":"event","date":"2015-09","basis":"publication","note":"Cover \"September 2015\". Checked in a current copy; an archive capture was not available, so earlier bytes are not authenticated."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"(First passage) NBER Working Paper 21547, September 2015 (version dated September 1, 2015). Bears on M09 (relation changed from articulates on 2026-10-12): the scarcity of Singularity writing in modern macroeconomics is narrower than the recent entry of machine intelligence into economics, and the same section discusses Simon's 1965 macroeconomic analysis of computerisation (third passage). It no longer holds M09's earliest-located role. (Second passage) NBER Working Paper 21547, September 2015 (version dated September 1, 2015); not pre-2015. (Third passage) Nordhaus's judgement (September 2015). Simon's book itself was not retrieved (a lead in the research pack).","evidenceIds":["evidence-ec-m09-11","evidence-ec-m09-12","evidence-ec-m09-35"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"(M09-11) The claim is about singularity in modern macroeconomics, after a substantive account of Simon; it cannot establish recent entry of machine intelligence generally. (M09-12) Describes economists' unfamiliarity with the material, not explicitly the field as new or neglected. (M09-35) A 2015 economist's account of earlier economic analysis of computerisation, not a description of the field as new; it bears on M09 as counter-context to an \"only recently\" reading.","tier":{"value":1,"basis":"NBER working paper (core venue); Nordhaus an economist."}}}},{"id":"ec-m09-calvano-et-al-2020-algorithmic-pricing-collusion-b","subject":"calvano-et-al-2020-algorithmic-pricing-collusion","predicate":"bears_on_proposition","object":{"kind":"entity","entityId":"recent-economics-engagement-with-ai"},"topic":"ideas","statement":"Calvano et al., “Artificial Intelligence, Algorithmic Pricing, and Collusion” (2020-10) bears on M09. The abstract shows a 2020 AER article whose subject is AI pricing algorithms in oligopoly.","time":{"kind":"event","date":"2020-10","basis":"publication","note":"AEA article page. Checked in a current copy; an archive capture was not available, so earlier bytes are not authenticated."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Abstract page only; a dated tier-1 text with AI as subject, relevant to the dated record under M09, not a self-description.","evidenceIds":["evidence-ec-m09-33"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"A dated data point, not a statement about the field's recency.","tier":{"value":1,"basis":"AER (core venue)."}}}},{"id":"lg-mx-nordeen-xai-team-2023","subject":"ross-nordeen","predicate":"held_role","object":{"kind":"entity","entityId":"xai"},"topic":"people","statement":"Ars Technica reported on 12 July 2023 that xAI's website listed Ross Nordeen among the members of its team at launch.","time":{"kind":"snapshot","date":"2023-07-12","note":"Snapshot of the roster shown on xAI's website at launch, as reported."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The record rests on reporting of xAI's launch web page, which was not retrieved directly. Roster membership is not a title or hiring date.","evidenceIds":["lg-ev-mx-nordeen-xai-team-2023"],"attributes":{"role":"Launch-team member"}},{"id":"lg-oai-karpathy-departure-2024","subject":"andrej-karpathy","predicate":"departure_announced","object":{"kind":"entity","entityId":"openai"},"topic":"people","statement":"Andrej Karpathy announced in a post dated 14 February 2024 (UTC) that he had left OpenAI the previous day.","time":{"kind":"event","date":"2024-02-14","note":"UTC date of the post (13 February in US Pacific time); the post says he left “yesterday”."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A first-person announcement. The post does not name his role; his own career account describes the 2023–2024 return (existing record).","evidenceIds":["lg-ev-oai-karpathy-departure-2024"],"attributes":{"role":"Not stated in the post"}},{"id":"lg-deal-google-windsurf-agreement-2025","subject":"google","predicate":"licensing_hiring_agreement","object":{"kind":"entity","entityId":"windsurf"},"topic":"institutions","statement":"Windsurf announced on 11 July 2025 an agreement with Google under which Varun, Douglas and some members of its R&D team would join Google; The Verge reported the same day that Google would take a non-exclusive licence to some of Windsurf's technology.","time":{"kind":"event","date":"2025-07-11","note":"Announcement date; signing and effective dates not stated."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Windsurf’s own post states the agreement and the hires but not the licence. The licence and Google’s lack of control or stake rest on The Verge’s reporting of the Google–Windsurf announcement; the Google spokesperson statement quoted there does not mention the licence. No Google primary statement was retrieved. Terms and payments are not recorded here.","evidenceIds":["lg-ev-deal-google-windsurf-agreement-licence","lg-ev-deal-google-windsurf-agreement-hiring","lg-ev-deal-google-windsurf-agreement-windsurf"],"attributes":{"activityStage":"announced"}},{"id":"lg-ant-cuellar-ltbt-trustee","subject":"mariano-florentino-cuellar","predicate":"held_role","object":{"kind":"entity","entityId":"anthropic-ltbt"},"topic":"governance","statement":"Anthropic's Trust page and its announcement of 4 August 2026 state that Mariano-Florentino (Tino) Cuéllar, President of the Carnegie Endowment for International Peace, joined its Long-Term Benefit Trust in January 2026.","time":{"kind":"interval","from":"2026-01","note":"Start month as stated in Anthropic's undated footnote and in its announcement of 4 August 2026 (\"since January 2026\"). His step-down is recorded separately; no end date is stated."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Documented roles only. The footnote is undated and was added after the original publication. The Company page retrieved on 15 October 2026 does not list him as a Trustee.","evidenceIds":["lg-ev-ant-cuellar-ltbt-trustee-1","lg-ev-ant-cuellar-ltbt-trustee-2","lg-ev-ant-cuellar-ltbt-trustee-company-page"],"attributes":{"role":"Trustee"}},{"id":"lg-oai-kc-cpo-2026","subject":"arvind-kc","predicate":"held_role","object":{"kind":"entity","entityId":"openai"},"topic":"people","statement":"OpenAI announced on 24 February 2026 that Arvind KC was joining as its Chief People Officer.","time":{"kind":"snapshot","date":"2026-02-24","note":"Announcement date. Publication date from OpenAI’s own news feed (openai.com/news/rss.xml, retrieved 2026-10-15); the reader-relay text does not carry it."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An announced appointment; no start date is given.","evidenceIds":["lg-ev-oai-kc-cpo-2026"],"attributes":{"activityStage":"announced","role":"Chief People Officer"}},{"id":"lg-cn-junyang-lin-qwen-departure-2026","subject":"junyang-lin","predicate":"departure_announced","object":{"kind":"entity","entityId":"qwen-team"},"topic":"people","statement":"Junyang Lin posted on X on 3 March 2026 (UTC; 4 March in Beijing) that he was stepping down from Qwen; Caixin Global reported on 4 March 2026 that he had resigned as a technical leader of Alibaba’s Qwen model team.","time":{"kind":"event","date":"2026-03-03","note":"Post timestamp 2026-03-03T16:11Z (4 March in Beijing). Announcement date, not a last working day."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The post gives no role, reason or next step; the role title rests on Caixin’s reporting. Reported internal Alibaba communications and further departures are not recorded here.","evidenceIds":["lg-ev-cn-junyang-lin-qwen-departure-2026","lg-ev-cn-junyang-lin-qwen-departure-2026-caixin"],"attributes":{"role":"technical leader, Qwen model team (as reported by Caixin)"}},{"id":"cmp-fsf-v2-0-d6-discretion","subject":"fsf-v2-0","predicate":"states_provision","object":{"kind":"literal","value":"Other exceptions or discretion"},"topic":"governance","statement":"FSF · v2.0 states a provision on the comparison question “Other exceptions or discretion”.","time":{"kind":"event","date":"2025-02-04","basis":"publication","note":"Comparison cell for this edition, dated by the edition’s publication date."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Other stated room for judgement: early warning evaluations may be run more frequently or alert thresholds adjusted (PDF p. 3). 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-fsf-v2-0-d6-discretion-1"],"attributes":{"provision":{"kind":"exception","key":"comparison-d6-discretion","facet":{"id":"D6.discretion","status":"stated"}}}},{"id":"ap-eliezer-yudkowsky-sl4-2000-08-21-yudkowsky-0008-0046","subject":"eliezer-yudkowsky","predicate":"authored","object":{"kind":"entity","entityId":"sl4-2000-08-21-yudkowsky-0008-0046"},"topic":"ideas","statement":"The mailing-list message “Re: [SL4] Employment vs. Singularity” (2000-08-21) is sent under the name Eliezer S. Yudkowsky.","time":{"kind":"event","date":"2000-08-21","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-sl4-2000-08-21-yudkowsky-0008-0046"]},{"id":"ap-eliezer-yudkowsky-lw-2022-six-dimensions-of-operational-adequacy-in","subject":"eliezer-yudkowsky","predicate":"authored","object":{"kind":"entity","entityId":"lw-2022-six-dimensions-of-operational-adequacy-in"},"topic":"ideas","statement":"“Six Dimensions of Operational Adequacy in AGI Projects” (2022-05-30) credits Eliezer Yudkowsky as its author.","time":{"kind":"event","date":"2022-05-30","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 authorship of this expression; it says nothing about priority for the ideas discussed.","evidenceIds":["evidence-ap-eliezer-yudkowsky-lw-2022-six-dimensions-of-operational-adequacy-in"]},{"id":"cmp-rsp-v2-0-d5-cadence","subject":"rsp-v2-0","predicate":"states_provision","object":{"kind":"literal","value":"Evaluation timing"},"topic":"governance","statement":"RSP · v2.0 states a provision on the comparison question “Evaluation timing”.","time":{"kind":"event","date":"2024-10-15","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-v2-0-d5-cadence-1"],"attributes":{"provision":{"kind":"evaluation_timing","key":"comparison-d5-cadence","facet":{"id":"D5.cadence","status":"stated"}}}},{"id":"cmp-rsp-v2-1-d1-other","subject":"rsp-v2-1","predicate":"states_provision","object":{"kind":"literal","value":"Other named domains"},"topic":"governance","statement":"RSP · v2.1 states a provision on the comparison question “Other named domains”, in a weaker form than the question’s criterion.","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":"Stated, partly: the passage has what the question’s criterion asks for, but in a weaker form than the criterion’s own words. Named as a possible future risk, not a tracked domain. Misaligned goals appear in the AI R&D-4 Required Safeguards, under which Anthropic says it will develop an affirmative case identifying the most immediate risks arising when models pursue misaligned goals (PDF p. 8). 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-d1-other-1"],"attributes":{"provision":{"kind":"risk_domain","key":"comparison-d1-other","facet":{"id":"D1.other","status":"partly"}}}},{"id":"fl-eu-aia-annex13-criteria","subject":"eu-ai-act-2024","predicate":"states_provision","object":{"kind":"literal","value":"Annex XIII criteria for designating models with systemic risk"},"topic":"governance","statement":"Annex XIII lists the criteria the Commission takes into account in deciding that a general-purpose AI model has capabilities or an impact equivalent to high impact capabilities: parameters, data set quality or size, training computation, input and output modalities, capability benchmarks and evaluations, internal-market impact by reach, and the number of registered end-users.","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). Annex XIII is applied through Article 51 in Chapter V, which 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-annex13-criteria-1","evidence-fl-eu-aia-annex13-criteria-2"],"attributes":{"provision":{"kind":"threshold","key":"aia-annex-xiii"}}},{"id":"cpo-ch-nvidia-openai-investment-intent","subject":"nvidia","predicate":"investment_announced","object":{"kind":"entity","entityId":"openai"},"topic":"capital","statement":"The joint OpenAI–NVIDIA release of 22 September 2025 stated that NVIDIA intended to invest up to $100 billion in OpenAI progressively as each gigawatt of NVIDIA systems is deployed.","time":{"kind":"event","date":"2025-09-22","basis":"announcement","note":"Date of the NVIDIA release."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A stated intention under a letter of intent, conditional on deployment (\"progressively as each gigawatt is deployed\"). It records no invested amount, no closing and no payment, and does not show the size of any single tranche. Reports of 2026 that the arrangement changed rest on press reporting and are not recorded here (see openQuestions).","evidenceIds":["evidence-cpo-ch-nvidia-openai-investment-intent-1"],"attributes":{"transactionType":"equity_investment","relationshipContext":"external","conditions":"progressively as each gigawatt is deployed","transactionStage":"letter_of_intent","amount":{"value":100000000000,"currency":"USD","bound":"up_to","scope":"transaction"}}},{"id":"cpo-orb-spacex-odc-plan","subject":"spacex","predicate":"announced_plan","object":{"kind":"entity","entityId":"spacex-orbital-data-center-system"},"topic":"compute","statement":"In its 2 February 2026 update on acquiring xAI, SpaceX stated a plan to launch a constellation of a million satellites operating as orbital data centres.","time":{"kind":"event","date":"2026-02-02","basis":"announcement","note":"Date of the announcement, as the source gives it."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A stated plan in a signed company letter. Its capacity figure is conditional arithmetic (a million tons per year of satellites at 100 kW per ton would add 100 GW a year), not a commitment or schedule. Nothing is deployed.","evidenceIds":["evidence-cpo-orb-spacex-odc-plan-1","evidence-cpo-orb-spacex-odc-plan-2"]},{"id":"kp-states-gpt-4-technical-report-2023-1","subject":"kp-gpt-4-technical-report-2023","predicate":"states_contribution","object":{"kind":"literal","value":"The report describes the development of GPT-4, which it presents as a large-scale multimodal model that accepts image and text inputs and produces text outputs."},"topic":"ideas","statement":"The report describes the development of GPT-4, which it presents as a large-scale multimodal model that accepts image and text inputs and produces text outputs.","time":{"kind":"event","date":"2023-03-15","basis":"publication","note":"Date of the arXiv version quoted, submitted 15 March 2023 (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-gpt-4-technical-report-2023-1-1"]},{"id":"kp-states-constitutional-ai-harmlessness-from-ai-2022-1","subject":"kp-constitutional-ai-harmlessness-from-ai-2022","predicate":"states_contribution","object":{"kind":"literal","value":"The paper describes a method it calls Constitutional AI (CAI) and its use to train a non-evasive and relatively harmless AI assistant without human feedback labels for harms."},"topic":"ideas","statement":"The paper describes a method it calls Constitutional AI (CAI) and its use to train a non-evasive and relatively harmless AI assistant without human feedback labels for harms.","time":{"kind":"event","date":"2022-12-15","basis":"publication","note":"Date of the arXiv version quoted, submitted 15 December 2022 (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-constitutional-ai-harmlessness-from-ai-2022-1-1"]},{"id":"kp-states-xgboost-scalable-tree-boosting-system-2016-1","subject":"kp-xgboost-scalable-tree-boosting-system-2016","predicate":"states_contribution","object":{"kind":"literal","value":"The paper describes XGBoost, a scalable machine learning system for tree boosting."},"topic":"ideas","statement":"The paper describes XGBoost, a scalable machine learning system for tree boosting.","time":{"kind":"event","date":"2016-03-09","basis":"publication","note":"Date of the arXiv version quoted, submitted 9 March 2016 (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-xgboost-scalable-tree-boosting-system-2016-1-1"]},{"id":"kp-cites-iem-2013-basic-ai-drives-2008","subject":"iem-2013","predicate":"cites","object":{"kind":"entity","entityId":"kp-basic-ai-drives-2008"},"topic":"ideas","statement":"Intelligence Explosion Microeconomics cites The Basic AI Drives.","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-basic-ai-drives-2008-1"]},{"id":"kp-cites-kimi-k2-report-triviaqa-large-scale-distantly-supervised-2017","subject":"kimi-k2-report","predicate":"cites","object":{"kind":"entity","entityId":"kp-triviaqa-large-scale-distantly-supervised-2017"},"topic":"ideas","statement":"Kimi K2: Open Agentic Intelligence cites TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension.","time":{"kind":"event","date":"2025-07-28","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-kimi-k2-report-triviaqa-large-scale-distantly-supervised-2017-1"]},{"id":"kp-cites-chatglm-report-llama-open-efficient-foundation-language-2023","subject":"chatglm-report","predicate":"cites","object":{"kind":"entity","entityId":"llama-paper"},"topic":"ideas","statement":"ChatGLM research history and techniques cites LLaMA: Open and Efficient Foundation Language Models.","time":{"kind":"event","date":"2024-06-18","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-chatglm-report-llama-open-efficient-foundation-language-2023-1"]},{"id":"kp-cites-self-critique-paper-self-consistency-improves-chain-thought-2022","subject":"self-critique-paper","predicate":"cites","object":{"kind":"entity","entityId":"kp-self-consistency-improves-chain-thought-2022"},"topic":"ideas","statement":"Self-critiquing models for assisting human evaluators cites Self-Consistency Improves Chain of Thought Reasoning in Language Models.","time":{"kind":"event","date":"2022-06-12","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-self-critique-paper-self-consistency-improves-chain-thought-2022-1"]},{"id":"kp-authored-niki-parmar-attention-all-you-need","subject":"niki-parmar","predicate":"authored","object":{"kind":"entity","entityId":"attention-all-you-need"},"topic":"ideas","statement":"Niki Parmar is a credited author of Attention Is All You Need.","time":{"kind":"event","date":"2017-06-12","basis":"publication","note":"Date of the work’s earliest recorded version, arXiv version 1, submitted 12 June 2017 (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-niki-parmar-attention-all-you-need-1"]},{"id":"kp-authored-alec-radford-ppo-paper","subject":"alec-radford","predicate":"authored","object":{"kind":"entity","entityId":"ppo-paper"},"topic":"ideas","statement":"Alec Radford is a credited author of Proximal Policy Optimization Algorithms.","time":{"kind":"event","date":"2017-07-20","basis":"publication","note":"Date of the work’s earliest recorded version, arXiv version 1, submitted 20 July 2017 (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-alec-radford-ppo-paper-1"]},{"id":"kp-authored-laurent-sifre-scaling-language-models-methods-analysis-2021","subject":"laurent-sifre","predicate":"authored","object":{"kind":"entity","entityId":"kp-scaling-language-models-methods-analysis-2021"},"topic":"ideas","statement":"Laurent Sifre is a credited author of Scaling Language Models: Methods, Analysis & Insights from Training Gopher.","time":{"kind":"event","date":"2021-12-08","basis":"publication","note":"Date of the work’s earliest recorded version, arXiv version 1, submitted 8 December 2021 (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-laurent-sifre-scaling-language-models-methods-analysis-2021-1"]},{"id":"th-term-llm-zhao-2023-def","subject":"th-zhao-2023-llm-survey","predicate":"discusses","object":{"kind":"entity","entityId":"th-term-large-language-model"},"topic":"ideas","statement":"Zhao et al. (arXiv v1, 2023) define large language models as Transformer-based language models with hundreds of billions or more parameters trained on massive text data.","time":{"kind":"event","date":"2023-03-31","note":"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 survey's working definition; its conclusion narrows its scope to models larger than 10B parameters, and p. 1 speaks of \"tens or hundreds of billions\". Other texts use the phrase without this threshold.","evidenceIds":["evidence-th-term-llm-zhao-2023-def-1","evidence-th-term-llm-zhao-2023-def-2","evidence-th-term-llm-zhao-2023-def-3"],"attributes":{"technical":{"kind":"definition"}}},{"id":"c3-sensetime-xu-li-chair-ceo-2021","subject":"xu-li","predicate":"held_role","object":{"kind":"entity","entityId":"sensetime"},"topic":"people","statement":"SenseTime’s December 2021 prospectus identifies Dr. Xu Li as executive Chairman of the Board, executive Director and chief executive officer, appointed a Director on 15 December 2015.","time":{"kind":"snapshot","date":"2021-12-07","note":"Roles shown in the prospectus of 7 December 2021. The prospectus dates his appointment as a Director to 15 December 2015 and his re-designation as an executive Director to 23 August 2021."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Roles as the filing states them at its date; see the 2025 annual report record for a later snapshot.","evidenceIds":["c3-ev-sensetime-xu-li-chair-ceo-2021"],"attributes":{"role":"executive Chairman of our Board, executive Director and chief executive officer"}},{"id":"c3-stepfun-step1x-edit-report","subject":"stepfun","predicate":"published","object":{"kind":"entity","entityId":"step1x-edit-report"},"topic":"ideas","statement":"A StepFun team posted the Step1X-Edit technical report to arXiv on 24 April 2025 (version 1), describing the Step1X-Edit image-editing model.","time":{"kind":"event","date":"2025-04-24","basis":"publication","note":"Submission date of the cited arXiv version; not the model's release date and not the date of later versions."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The report is the developer's own description; performance and benchmark claims in it are out of scope. Credit to StepFun rests on the team line of the report, not on a separate statement of ownership.","evidenceIds":["c3-ev-stepfun-step1x-edit-report","c3-ev-stepfun-step1x-edit-report-2"]},{"id":"c3-governance-cac-genai-list-2026-08-count-registered","subject":"cn-cac-genai-service-list-2026-08","predicate":"lists_participants","object":{"kind":"literal","value":"133 registered generative AI applications or functions"},"topic":"governance","statement":"The August 2026 CAC list has 133 rows of registered generative AI applications or functions, numbered 599 to 731 (count made by the preparer).","time":{"kind":"snapshot","date":"2026-08","note":"Month of the listing as its title gives it. Filing (or registration) dates of individual rows are separate dates listed in the rows."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Count made by the preparer from the numbered rows of the retrieved PDF (pdftotext layout); a row may be miscounted where the table wraps, but the row numbers ran without a gap. Rows 730 and 731 are marked 补登 (backfilled), with registration dates of 1 May and 11 June 2026, so the 133 rows are the registrations this attachment lists, not all registrations made in July and August 2026. Registration (登记) is a separate procedure from filing (备案): the CAC's January 2026 announcement, cited here as context, describes it as applying to applications or functions that call filed models through an API or otherwise. Registration is not an assessment of safety, performance or compliance.","evidenceIds":["c3-ev-governance-cac-genai-list-2026-08-count-registered-1","c3-ev-governance-cac-genai-list-2026-08-count-registered-registration-announcement"],"attributes":{"listing":{"count":133,"countBasis":"counted"}}},{"id":"rd-reply-narayanan-kapoor-2024-ai-scaling-myths-to-aschenbrenner-2024-situational-awareness","subject":"narayanan-kapoor-2024-ai-scaling-myths","predicate":"cites","object":{"kind":"entity","entityId":"aschenbrenner-2024-situational-awareness"},"topic":"ideas","statement":"Narayanan and Kapoor's post of June 2024 addresses Aschenbrenner's essay of 2024, which it names, in its further-reading note.","time":{"kind":"event","date":"2024-06-27","basis":"publication","note":"Date of posting."},"derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A reply is not agreement; the record says only that the text answers the other.","evidenceIds":["evidence-rd-reply-narayanan-kapoor-2024-ai-scaling-myths-to-aschenbrenner-2024-situational-awareness-1","evidence-rd-reply-narayanan-kapoor-2024-ai-scaling-myths-to-aschenbrenner-2024-situational-awareness-2"]}],"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-xai-team-2023","title":"Musk announces new AI company that seeks to “understand the universe”","authors":["Benj Edwards"],"publisher":"Ars Technica","publicationDate":"2023-07-12","version":"Contemporaneous report and screenshot of the launch roster. The current x.ai company page has changed; original archive retrieval was unavailable.","canonicalUrl":"https://arstechnica.com/information-technology/2023/07/musk-announces-new-ai-company-that-seeks-to-understand-the-universe/","evidenceUrl":"https://arstechnica.com/information-technology/2023/07/musk-announces-new-ai-company-that-seeks-to-understand-the-universe/","language":"en","archivedUrl":"https://web.archive.org/web/20260211075334/https://arstechnica.com/information-technology/2023/07/musk-announces-new-ai-company-that-seeks-to-understand-the-universe/","archivedAt":"2026-02-11","accessedAt":"2026-09-18","reuse":{"attribution":"Benj Edwards, Musk announces new AI company that seeks to “understand the universe” (2023-07-12).","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-2-0","documentId":"rsp-v2-0","title":"Responsible Scaling Policy — v2.0","authors":["Anthropic"],"publisher":"Anthropic","publisherId":"anthropic","publicationDate":null,"version":"Responsible Scaling Policy v2.0; effective date 2024-10-15. v2.0 identity is supplied by the official archive and later dated changelogs; the cover is not numbered.","versionDate":"2024-10-15","canonicalUrl":"https://www-cdn.anthropic.com/616dee633636e5bd309cb73aed8622e80fe47839.pdf","evidenceUrl":"https://www-cdn.anthropic.com/616dee633636e5bd309cb73aed8622e80fe47839.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20241022154907/https://assets.anthropic.com/m/24a47b00f10301cd/original/Anthropic-Responsible-Scaling-Policy-2024-10-15.pdf","archivedAt":"2024-10-22","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v2.0.","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-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-fsf-2-0","documentId":"fsf-v2-0","title":"Frontier Safety Framework — v2.0","authors":["Google DeepMind"],"publisher":"Google DeepMind","publisherId":"google-deepmind","publicationDate":"2025-02-04","version":"Frontier Safety Framework v2.0; publication date 2025-02-04.","versionDate":"2025-02-04","canonicalUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/updating-the-frontier-safety-framework/Frontier%20Safety%20Framework%202.0.pdf","evidenceUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/updating-the-frontier-safety-framework/Frontier%20Safety%20Framework%202.0.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20250205011727/https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/updating-the-frontier-safety-framework/Frontier%20Safety%20Framework%202.0.pdf","archivedAt":"2025-02-05","accessedAt":"2026-09-18","reuse":{"attribution":"Google DeepMind, Frontier Safety Framework — v2.0.","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-attention-all-you-need-v1","documentId":"attention-all-you-need","title":"Attention Is All You Need","authors":["Ashish Vaswani","Noam Shazeer","Niki Parmar","Jakob Uszkoreit","Llion Jones","Aidan N. Gomez","Lukasz Kaiser","Illia Polosukhin"],"publisher":"arXiv","publicationDate":"2017-06-12","version":"arXiv:1706.03762v1; first-submission date from version history. Later revisions are not substituted.","versionDate":"2017-06-12","canonicalUrl":"https://arxiv.org/abs/1706.03762","evidenceUrl":"https://arxiv.org/pdf/1706.03762v1","language":"en","accessedAt":"2026-09-20","reuse":{"attribution":"Ashish Vaswani; Noam Shazeer; Niki Parmar; Jakob Uszkoreit; Llion Jones; Aidan N. Gomez; Lukasz Kaiser; Illia Polosukhin, Attention Is All You Need (2017-06-12).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; no full source, personal contact details or dataset imported."}},{"id":"source-ppo-paper-v1","documentId":"ppo-paper","title":"Proximal Policy Optimization Algorithms","authors":["John Schulman","Filip Wolski","Prafulla Dhariwal","Alec Radford","Oleg Klimov"],"publisher":"arXiv","publicationDate":"2017-07-20","version":"arXiv:1707.06347v1; first-submission date from version history. Later revisions are not substituted.","versionDate":"2017-07-20","canonicalUrl":"https://arxiv.org/abs/1707.06347","evidenceUrl":"https://arxiv.org/pdf/1707.06347v1","language":"en","accessedAt":"2026-09-20","reuse":{"attribution":"John Schulman; Filip Wolski; Prafulla Dhariwal; Alec Radford; Oleg Klimov, Proximal Policy Optimization Algorithms (2017-07-20).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; no full source, personal contact details or dataset imported."}},{"id":"source-self-critique-paper-v1","documentId":"self-critique-paper","title":"Self-critiquing models for assisting human evaluators","authors":["William Saunders","Catherine Yeh","Jeff Wu","Steven Bills","Long Ouyang","Jonathan Ward","Jan Leike"],"publisher":"arXiv","publicationDate":"2022-06-12","version":"arXiv:2206.05802v1; first-submission date from version history. Later revisions are not substituted.","versionDate":"2022-06-12","canonicalUrl":"https://arxiv.org/abs/2206.05802","evidenceUrl":"https://arxiv.org/pdf/2206.05802v1","language":"en","accessedAt":"2026-09-25","reuse":{"attribution":"William Saunders; Catherine Yeh; Jeff Wu; Steven Bills; Long Ouyang; Jonathan Ward; Jan Leike, Self-critiquing models for assisting human evaluators.","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-cn-chatglm","title":"ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools","authors":["Team GLM"],"publisher":"Team GLM","publicationDate":"2024-06-18","version":"Source language: English (en). arXiv:2406.12793v1; joint Zhipu AI / Tsinghua University affiliation.","canonicalUrl":"https://arxiv.org/html/2406.12793v1","evidenceUrl":"https://arxiv.org/html/2406.12793v1","language":"en","accessedAt":"2026-09-29","reuse":{"attribution":"Team GLM, ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools (2024-06-18).","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-ca-sb53-chaptered","documentId":"california-sb-53","title":"SB-53 Artificial intelligence models: large developers. (2025-2026) — bill text, chaptered version (Chapter 138, Statutes of 2025)","authors":["California Legislature"],"publisher":"California State Legislature","publisherId":"california-legislature","publicationDate":"2025-09-29","version":"Chaptered version dated 09/29/25, inspected live 2026-10-04; quote also matched in the 2025-09-30 capture. Publisher as printed on the inspected expression: California Legislature (California Legislative Information, Office of Legislative Counsel).","canonicalUrl":"https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB53","evidenceUrl":"https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB53","language":"en","archivedUrl":"https://web.archive.org/web/20250930154105/https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB53","archivedAt":"2025-09-30","accessedAt":"2026-10-04","reuse":{"attribution":"California Legislature, SB-53 Artificial intelligence models: large developers. (2025-2026) — bill text, chaptered version (Chapter 138, Statutes of 2025).","licence":"Official government 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-ca-bpc-22757-15","title":"Business and Professions Code § 22757.15 (codified section, California Law)","authors":["California Legislature"],"publisher":"California State Legislature","publisherId":"california-legislature","publicationDate":null,"version":"Live codified section retrieved 2026-10-04; section text identical to the chaptered bill text. Publisher as printed on the inspected expression: California Legislature (California Legislative Information, Office of Legislative Counsel).","canonicalUrl":"https://leginfo.legislature.ca.gov/faces/codes_displaySection.xhtml?lawCode=BPC&sectionNum=22757.15.","evidenceUrl":"https://leginfo.legislature.ca.gov/faces/codes_displaySection.xhtml?lawCode=BPC&sectionNum=22757.15.","language":"en","archivedUrl":"https://web.archive.org/web/20260130074817/https://leginfo.legislature.ca.gov/faces/codes_displaySection.xhtml?lawCode=BPC&sectionNum=22757.15","archivedAt":"2026-01-30","accessedAt":"2026-10-04","reuse":{"attribution":"California Legislature, Business and Professions Code § 22757.15 (codified section, California Law).","licence":"Official government 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-openai-group-pbc-certificate","documentId":"openai-group-pbc-certificate-2025","title":"Certificate of Incorporation of OpenAI Group PBC","authors":["OpenAI Group PBC"],"publisher":"OpenAI Group PBC (filed with the Delaware Division of Corporations)","publisherId":"openai-group-pbc","publicationDate":"2025-10-28","version":"Filed 2025-10-28 at 08:07 AM according to the Division of Corporations stamp (file number 10381551); 30-page scan with an OCR text layer, downloaded from DocumentCloud on 2026-10-04 and byte-identical (SHA-1) to the Internet Archive captures from 2025-10-29. Quotes were read from the page images because the OCR layer contains recognition errors.","canonicalUrl":"https://s3.documentcloud.org/documents/26205026/openai-pbc-articles-of-incorporation.pdf","evidenceUrl":"https://s3.documentcloud.org/documents/26205026/openai-pbc-articles-of-incorporation.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20251029204930/https://s3.documentcloud.org/documents/26205026/openai-pbc-articles-of-incorporation.pdf","archivedAt":"2025-10-29","accessedAt":"2026-10-04","reuse":{"attribution":"OpenAI Group PBC, Certificate of Incorporation (filed 28 October 2025), copy hosted by DocumentCloud.","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-ltbt-2023","documentId":"anthropic-ltbt-announcement-2023","title":"The Long-Term Benefit Trust","authors":["Anthropic"],"publisher":"Anthropic","publisherId":"anthropic","publicationDate":"2023-09-19","version":"Web page dated Sep 19, 2023, inspected live 2026-10-04 at its current address; quotes also matched in the Internet Archive capture of 2023-09-25 at the original address. The live page adds an undated footnote on later changes in the Trust’s membership.","canonicalUrl":"https://www.anthropic.com/news/the-long-term-benefit-trust","evidenceUrl":"https://www.anthropic.com/news/the-long-term-benefit-trust","language":"en","archivedUrl":"https://web.archive.org/web/20230925170803/https://www.anthropic.com/index/the-long-term-benefit-trust","archivedAt":"2023-09-25","accessedAt":"2026-10-04","reuse":{"attribution":"Anthropic, The Long-Term Benefit Trust (19 September 2023).","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-agi-ruin-2022","documentId":"agi-ruin-2022","title":"AGI Ruin: A List of Lethalities","authors":["Eliezer Yudkowsky"],"publisher":"LessWrong and Alignment Forum","publicationDate":"2022-06-05","version":"Text checked in the post’s first stored version through the LessWrong GraphQL API on 2026-10-06.","canonicalUrl":"https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a-list-of-lethalities","evidenceUrl":"https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a-list-of-lethalities","language":"en","archivedUrl":"https://web.archive.org/web/20220605225854/https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a-list-of-lethalities","archivedAt":"2022-06-05","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, \"AGI Ruin: A List of Lethalities\", LessWrong and Alignment Forum, 2022-06-05.","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-lw-2022-six-dimensions-of-operational-adequacy-in","documentId":"lw-2022-six-dimensions-of-operational-adequacy-in","title":"Six Dimensions of Operational Adequacy in AGI Projects","authors":["Eliezer Yudkowsky"],"publisher":"LessWrong","publicationDate":"2022-05-30","version":"Post body retrieved through the LessWrong GraphQL API on 2026-10-06 (SHA-256 30f88ccb952602dd97747a4ce6188fd120a8b51ce44b8acc73f80411f634f223), content version 1.1.1; quotations also compared with the posting-day version 1.0.0.","canonicalUrl":"https://www.lesswrong.com/posts/keiYkaeoLHoKK4LYA/six-dimensions-of-operational-adequacy-in-agi-projects","evidenceUrl":"https://www.lesswrong.com/posts/keiYkaeoLHoKK4LYA/six-dimensions-of-operational-adequacy-in-agi-projects","language":"en","archivedUrl":"https://web.archive.org/web/20261006005922/https://www.lesswrong.com/posts/keiYkaeoLHoKK4LYA/six-dimensions-of-operational-adequacy-in-agi-projects","archivedAt":"2026-10-06","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, “Six Dimensions of Operational Adequacy in AGI Projects”, LessWrong, 2022-05-30.","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":"source-sl4-2000-08-21-yudkowsky-0008-0046","documentId":"sl4-2000-08-21-yudkowsky-0008-0046","title":"Re: [SL4] Employment vs. Singularity (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2000-08-21","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 aa21839afaeb0cb5f7e88ecb0ddffee6c7593fb3744c67a0dcab444452eb9b2a); 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/0008/0046.html","evidenceUrl":"http://sl4.org/archive/0008/0046.html","language":"en","archivedUrl":"https://web.archive.org/web/20230607061713/http://sl4.org/archive/0008/0046.html","archivedAt":"2023-06-07","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: [SL4] Employment vs. Singularity”, SL4 mailing list, 2000-08-21.","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-clancy-besiroglu-2023-great-inflection","documentId":"clancy-besiroglu-2023-great-inflection","title":"The Great Inflection? A Debate About AI and Explosive Growth","authors":["Matt Clancy","Tamay Besiroglu"],"publisher":"Asterisk magazine, issue 03","publicationDate":"2023-06","version":"Wayback capture of asteriskmag.com/issues/03/the-great-inflection-a-debate-about-ai-and-explosive-growth. Capture retrieved 2026-10-08 (SHA-256 e8b29f97be40d49c282cac8d279cc008644b822a4607533999157375d01753d1). Published version: Clancy, M., & Besiroglu, T. (2023, June). The great inflection? A debate about AI and explosive growth. Asterisk, (3). Magazine debate.","canonicalUrl":"https://asteriskmag.com/issues/03/the-great-inflection-a-debate-about-ai-and-explosive-growth","evidenceUrl":"https://web.archive.org/web/20231223213358/https://asteriskmag.com/issues/03/the-great-inflection-a-debate-about-ai-and-explosive-growth","language":"en","archivedUrl":"https://web.archive.org/web/20231223213358/https://asteriskmag.com/issues/03/the-great-inflection-a-debate-about-ai-and-explosive-growth","archivedAt":"2023-12-23","accessedAt":"2026-10-08","reuse":{"attribution":"Matt Clancy, Tamay Besiroglu, “The Great Inflection? A Debate About AI and Explosive Growth”, Asterisk magazine, issue 03, 2023-06.","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-nordhaus-2015-nber","documentId":"nordhaus-2015-economic-singularity","title":"Are We Approaching an Economic Singularity? Information Technology and the Future of Economic Growth","authors":["William D. Nordhaus"],"publisher":"NBER Working Paper 21547","publicationDate":"2015-09","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 cb481e5e99ef9bad8575d9150bcc8464eda4904722cc5048389ab6159858b062. Journal version not retrieved. Nordhaus, W. D. (2021). Are we approaching an economic singularity? Information technology and the future of economic growth. American Economic Journal: Macroeconomics, 13(1), 299–332. https://doi.org/10.1257/mac.20170105","canonicalUrl":"https://www.nber.org/system/files/working_papers/w21547/w21547.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w21547/w21547.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"William D. Nordhaus, “Are We Approaching an Economic Singularity? Information Technology and the Future of Economic Growth”, NBER Working Paper 21547, 2015-09.","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-acemoglu-restrepo-2016-race-between-machine-and-man","documentId":"acemoglu-restrepo-2016-race-between-machine-and-man","title":"The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment","authors":["Daron Acemoglu","Pascual Restrepo"],"publisher":"NBER Working Paper 22252","publicationDate":"2016-05","version":"Full text (revised version). Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 441d01202afd56ef8002fc24ffc2beb51191741c0b5accb11d2534620dd616b7. Journal version not retrieved. Acemoglu, D., & Restrepo, P. (2018). The race between man and machine: Implications of technology for growth, factor shares, and employment. American Economic Review, 108(6), 1488–1542. https://doi.org/10.1257/aer.20160696","canonicalUrl":"https://www.nber.org/system/files/working_papers/w22252/w22252.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w22252/w22252.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Daron Acemoglu, Pascual Restrepo, “The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment”, NBER Working Paper 22252, 2016-05.","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-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-calvano-et-al-2020-algorithmic-pricing-collusion","documentId":"calvano-et-al-2020-algorithmic-pricing-collusion","title":"Artificial Intelligence, Algorithmic Pricing, and Collusion","authors":["Emilio Calvano","Giacomo Calzolari","Vincenzo Denicolò","Sergio Pastorello"],"publisher":"American Economic Review 110(10), 2020, pp. 3267–97","publicationDate":"2020-10","version":"Abstract page only. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 48c037a5cf242fca6ad8f4c36d38f34e248b31d856fb6b1d9475eed61f5925be. Full text not retrieved; only the AEA abstract page. Calvano, E., Calzolari, G., Denicolò, V., & Pastorello, S. (2020). Artificial intelligence, algorithmic pricing, and collusion. American Economic Review, 110(10), 3267–3297. https://doi.org/10.1257/aer.20190623","canonicalUrl":"https://www.aeaweb.org/articles?id=10.1257/aer.20190623","evidenceUrl":"https://www.aeaweb.org/articles?id=10.1257/aer.20190623","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Emilio Calvano, Giacomo Calzolari, Vincenzo Denicolò, Sergio Pastorello, “Artificial Intelligence, Algorithmic Pricing, and Collusion”, American Economic Review 110(10), 2020, pp. 3267–97, 2020-10.","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-ant-company-2026","title":"Company (web page, undated; retrieved 2026-10-15)","authors":["Anthropic"],"publisher":"Anthropic","publisherId":"anthropic","publicationDate":null,"version":"Undated page, retrieved on 2026-10-15 (SHA-256 7a67ff17b2d81e48de09d746e810f49620ee2acfec6ec8ef6006f01044942827) directly from the publisher. Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://www.anthropic.com/company","evidenceUrl":"https://www.anthropic.com/company","language":"en","archivedUrl":"https://web.archive.org/web/20261015200453/https://www.anthropic.com/company","archivedAt":"2026-10-15","accessedAt":"2026-10-15","reuse":{"attribution":"Anthropic, “Company (web page, undated; retrieved 2026-10-15)”.","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":"lg-src-ant-cuellar-2026","title":"Mariano-Florentino (Tino) Cuéllar to join Anthropic as Chief Global Affairs Officer","authors":["Anthropic"],"publisher":"Anthropic","publisherId":"anthropic","publicationDate":"2026-08-04","version":"Dated page, retrieved on 2026-10-15 (SHA-256 1cc1f4cfda28d9c5c29e8ee583d8fe881600649b5826115286f3243eda3cbc59) directly from the publisher.","canonicalUrl":"https://www.anthropic.com/news/tino-cuellar","evidenceUrl":"https://www.anthropic.com/news/tino-cuellar","language":"en","archivedUrl":"https://web.archive.org/web/20260915005327/https://www.anthropic.com/news/tino-cuellar","archivedAt":"2026-09-15","accessedAt":"2026-10-15","reuse":{"attribution":"Anthropic, “Mariano-Florentino (Tino) Cuéllar to join Anthropic as Chief Global Affairs Officer”, 2026-08-04.","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-caixin-lin-2026","title":"Alibaba’s AI Model Chief Resigns","authors":["Qu Yunxu"],"publisher":"Caixin Global","publicationDate":"2026-03-04","version":"Dated page, retrieved on 2026-10-15 (SHA-256 9195fc001f5bc0b3e8fbd8edee4e8c5d767207c68eb6b410160750740e01d789) directly from the publisher.","canonicalUrl":"https://www.caixinglobal.com/2026-03-04/alibabas-ai-model-chief-resigns-102419448.html","evidenceUrl":"https://www.caixinglobal.com/2026-03-04/alibabas-ai-model-chief-resigns-102419448.html","language":"en","archivedUrl":"https://web.archive.org/web/20260925104822/https://www.caixinglobal.com/2026-03-04/alibabas-ai-model-chief-resigns-102419448.html","archivedAt":"2026-09-25","accessedAt":"2026-10-15","reuse":{"attribution":"Qu Yunxu, “Alibaba’s AI Model Chief Resigns”, 2026-03-04.","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-lin-junyang-x-2026","title":"Post by Junyang Lin (@JustinLin610), 3 March 2026","authors":["Junyang Lin"],"publisher":"X","publicationDate":"2026-03-03","version":"Dated post, retrieved on 2026-10-15 (SHA-256 4d41209c4278020b815b42ec28450fed8ab73d2da031acde0170c9524ca96ef8) through X’s own syndication endpoint for the post’s text, which truncates long notes.","canonicalUrl":"https://x.com/JustinLin610/status/2028865835373359513","evidenceUrl":"https://x.com/JustinLin610/status/2028865835373359513","language":"en","archivedUrl":"https://web.archive.org/web/20260303161124/https://twitter.com/JustinLin610/status/2028865835373359513","archivedAt":"2026-03-03","accessedAt":"2026-10-15","reuse":{"attribution":"Junyang Lin, “Post by Junyang Lin (@JustinLin610), 3 March 2026”, 2026-03-03.","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-deal-verge-windsurf-2025","title":"OpenAI’s Windsurf deal is off — and Windsurf’s CEO is going to Google","authors":["Hayden Field"],"publisher":"The Verge","publicationDate":"2025-07-11","version":"Dated page, retrieved on 2026-10-15 (SHA-256 adad993dc5cfd0bef51776f1a138314fc223c803d9ac0155a1e9367316399e59) directly from the publisher.","canonicalUrl":"https://www.theverge.com/openai/705999/google-windsurf-ceo-openai","evidenceUrl":"https://www.theverge.com/openai/705999/google-windsurf-ceo-openai","language":"en","archivedUrl":"https://web.archive.org/web/20261013220322/https://www.theverge.com/openai/705999/google-windsurf-ceo-openai","archivedAt":"2026-10-13","accessedAt":"2026-10-15","reuse":{"attribution":"Hayden Field, “OpenAI’s Windsurf deal is off — and Windsurf’s CEO is going to Google”, 2025-07-11.","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-deal-windsurf-next-stage-2025","title":"The Next Stage of Windsurf","authors":["Windsurf"],"publisher":"Windsurf","publisherId":"windsurf","publicationDate":"2025-07-11","version":"Dated page, retrieved on 2026-10-15 (SHA-256 d6787157ec7693af21f56d8242f8ecf8fad3d88d466064a6fefed222fe428ed0) 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.","canonicalUrl":"https://devin.ai/blog/windsurfs-next-stage","evidenceUrl":"https://windsurf.com/blog/windsurfs-next-stage","language":"en","archivedUrl":"https://web.archive.org/web/20260511121317/https://windsurf.com/blog/windsurfs-next-stage","archivedAt":"2026-05-11","accessedAt":"2026-10-15","reuse":{"attribution":"Windsurf, “The Next Stage of Windsurf”, 2025-07-11.","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-karpathy-x-2024","title":"Post of 14 February 2024 (UTC) on leaving OpenAI","authors":["Andrej Karpathy"],"publisher":"X","publicationDate":"2024-02-14","version":"Dated post, retrieved on 2026-10-15 (SHA-256 a5fa8a96eb8b3d8aaa53ef07061ddbd43521603716b2c6462eed782239697def) 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/karpathy/status/1757600075281547344","evidenceUrl":"https://x.com/karpathy/status/1757600075281547344","language":"en","archivedUrl":"https://web.archive.org/web/20261012084753/https://x.com/karpathy/status/1757600075281547344","archivedAt":"2026-10-12","accessedAt":"2026-10-15","reuse":{"attribution":"Andrej Karpathy, “Post of 14 February 2024 (UTC) on leaving OpenAI”, 2024-02-14.","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":"lg-src-oai-kc-2026","title":"Arvind KC appointed Chief People Officer","authors":["OpenAI"],"publisher":"OpenAI","publisherId":"openai","publicationDate":"2026-02-24","version":"Dated page, retrieved on 2026-10-15 (SHA-256 358783aa54c4680a5dca2e7f405e00f56f76f891a348b6eab046c6568dbe3b62) 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.","canonicalUrl":"https://openai.com/index/arvind-kc-chief-people-officer/","evidenceUrl":"https://openai.com/index/arvind-kc-chief-people-officer/","language":"en","archivedUrl":"https://web.archive.org/web/20260801041640/https://openai.com/index/arvind-kc-chief-people-officer/","archivedAt":"2026-08-01","accessedAt":"2026-10-15","reuse":{"attribution":"OpenAI, “Arvind KC appointed Chief People Officer”, 2026-02-24.","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":"c2-src-kimi-k2-pdf-v1","documentId":"kimi-k2-report","title":"Kimi K2: Open Agentic Intelligence (technical report, v1)","authors":["Kimi Team"],"publisher":"arXiv","publicationDate":"2025-07-28","version":"Dated document, retrieved on 2026-10-27 (SHA-256 b838917cf0a475d2e1b2f1ba4d0b4c91b6f6c8be95f15aa0d4f3ea76e4ef60f5) directly from the publisher. Version: 2507.20534v1. PDF of version 1; later versions not inspected.","canonicalUrl":"https://arxiv.org/pdf/2507.20534v1","evidenceUrl":"https://arxiv.org/pdf/2507.20534v1","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"Kimi Team, “Kimi K2: Open Agentic Intelligence (technical report, v1)”, 2025-07-28.","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":"source-cpo-orb-spacex-xai","documentId":"spacex-xai-update-2026","title":"xAI joins SpaceX to Accelerate Humanity’s Future","authors":["SpaceX"],"publisher":"SpaceX (spacex.com updates, via its content API)","publicationDate":"2026-02-02","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 84d2b34242c8ac9bbb47bf61dd0c53e4faed7df6604bc0121da21782f86676fc). spacex.com/updates serves a JavaScript shell; the text was retrieved from the site's own content API (content.spacex.com/api/spacex-website/updates), entry \"xai-joins-spacex\", date 2026-02-02, signed \"Elon\". The dataset's source-xm-spacex is the x.ai copy of the same announcement, a separate expression on another host (x.ai refuses retrieval from here); its evidence is a paraphrase, so no dataset quotation counts against this budget. Retrieved again at admission on 2026-10-30 (SHA-256 84d2b34242c8ac9bbb47bf61dd0c53e4faed7df6604bc0121da21782f86676fc, the same bytes). Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://content.spacex.com/api/spacex-website/updates","evidenceUrl":"https://content.spacex.com/api/spacex-website/updates","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"SpaceX, “xAI joins SpaceX to Accelerate Humanity’s Future”, 2026-02-02.","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-cpo-ch-openai-nvidia-2025","title":"OpenAI and NVIDIA Announce Strategic Partnership to Deploy 10 Gigawatts of NVIDIA Systems","authors":["NVIDIA (NVIDIA Newsroom)"],"publisher":"NVIDIA (NVIDIA Newsroom)","publicationDate":"2025-09-22","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 f508dd752cfd8f1b288b2d447c428d6a046d0ea140ba8187f3937a2d5fa699ac). Press release as served on 2026-10-28; dated September 22, 2025. OpenAI's own post on the same announcement could not be retrieved (openai.com refuses direct retrieval from this environment); the cross-family checker should open it or its capture. Retrieved again at admission on 2026-10-30 (SHA-256 ffa6da3512fd4bc20607cc222d5a9cabc868805dc17fb6764d23601774e8d275). Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://nvidianews.nvidia.com/news/openai-and-nvidia-announce-strategic-partnership-to-deploy-10gw-of-nvidia-systems","evidenceUrl":"https://nvidianews.nvidia.com/news/openai-and-nvidia-announce-strategic-partnership-to-deploy-10gw-of-nvidia-systems","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"NVIDIA (NVIDIA Newsroom), “OpenAI and NVIDIA Announce Strategic Partnership to Deploy 10 Gigawatts of NVIDIA Systems”, 2025-09-22.","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-gpt-4-technical-report-2023-v1","documentId":"kp-gpt-4-technical-report-2023","title":"GPT-4 Technical Report","authors":["OpenAI"],"publisher":"arXiv","publicationDate":"2023-03-15","version":"arXiv:2303.08774v1, submitted 15 March 2023 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 053056a10114d22e4c47b6b5be25e54c320b5f1beeae7466e8638dac0f5f5f66). The first page prints the collective byline “OpenAI”; contributors listed elsewhere in the text are not recorded as authors.","versionDate":"2023-03-15","canonicalUrl":"https://arxiv.org/abs/2303.08774v1","evidenceUrl":"https://arxiv.org/pdf/2303.08774v1","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"OpenAI, “GPT-4 Technical Report”, 2023-03-15.","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":"source-kp-constitutional-ai-harmlessness-from-ai-2022-v1","documentId":"kp-constitutional-ai-harmlessness-from-ai-2022","title":"Constitutional AI: Harmlessness from AI Feedback","authors":["Yuntao Bai","Saurav Kadavath","Sandipan Kundu","Amanda Askell","Jackson Kernion","Andy Jones","Anna Chen","Anna Goldie","Azalia Mirhoseini","Cameron McKinnon","Carol Chen","Catherine Olsson","Christopher Olah","Danny Hernandez","Dawn Drain","Deep Ganguli","Dustin Li","Eli Tran-Johnson","Ethan Perez","Jamie Kerr","Jared Mueller","Jeffrey Ladish","Joshua Landau","Kamal Ndousse","Kamile Lukosuite","Liane Lovitt","Michael Sellitto","Nelson Elhage","Nicholas Schiefer","Noemi Mercado","Nova DasSarma","Robert Lasenby","Robin Larson","Sam Ringer","Scott Johnston","Shauna Kravec","Sheer El Showk","Stanislav Fort","Tamera Lanham","Timothy Telleen-Lawton","Tom Conerly","Tom Henighan","Tristan Hume","Samuel R. Bowman","Zac Hatfield-Dodds","Ben Mann","Dario Amodei","Nicholas Joseph","Sam McCandlish","Tom Brown","Jared Kaplan"],"publisher":"arXiv","publicationDate":"2022-12-15","version":"arXiv:2212.08073v1, submitted 15 December 2022 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 9a456a07ad346e3372f9867d346f69f5b0f68b4c65f060aca0b8a13fa9d98e83). The author list is the byline as printed on the first page.","versionDate":"2022-12-15","canonicalUrl":"https://arxiv.org/abs/2212.08073v1","evidenceUrl":"https://arxiv.org/pdf/2212.08073v1","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Yuntao Bai and others, “Constitutional AI: Harmlessness from AI Feedback”, 2022-12-15.","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":"source-kp-scaling-language-models-methods-analysis-2021-v1","documentId":"kp-scaling-language-models-methods-analysis-2021","title":"Scaling Language Models: Methods, Analysis & Insights from Training Gopher","authors":["Jack W. Rae","Sebastian Borgeaud","Trevor Cai","Jordan Hoffmann","John Aslanides","Sarah Henderson","Roman Ring","Susannah Young","Eliza Rutherford","Tom Hennigan","Jacob Menick","Albin Cassirer","Richard Powell","Lisa Anne Hendricks","Maribeth Rauh","Po-Sen Huang","Amelia Glaese","Johannes Welbl","Sumanth Dathathri","Saffron Huang","Jonathan Uesato","John Mellor","Irina Higgins","Antonia Creswell","Nat McAleese","Amy Wu","Erich Elsen","Siddhant Jayakumar","Elena Buchatskaya","David Budden","Esme Sutherland","Karen Simonyan","Michela Paganini","Laurent Sifre","Lena Martens","Xiang Lorraine Li","Adhiguna Kuncoro","Aida Nematzadeh","Elena Gribovskaya","Domenic Donato","Angeliki Lazaridou","Arthur Mensch","Jean-Baptiste Lespiau","Maria Tsimpoukelli","Nikolai Grigorev","Doug Fritz","Thibault Sottiaux","Mantas Pajarskas","Toby Pohlen","Zhitao Gong","Daniel Toyama","Yujia Li","Tayfun Terzi","Vladimir Mikulik","Igor Babuschkin","Aidan Clark","Diego de Las Casas","Aurelia Guy","Chris Jones","James Bradbury","Matthew Johnson","Blake Hechtman","Laura Weidinger","Iason Gabriel","William Isaac","Ed Lockhart","Simon Osindero","Laura Rimell","Chris Dyer","Oriol Vinyals","Kareem Ayoub","Jeff Stanway","Lorrayne Bennett","Demis Hassabis","Koray Kavukcuoglu","Geoffrey Irving"],"publisher":"arXiv","publicationDate":"2021-12-08","version":"arXiv:2112.11446v1, submitted 8 December 2021 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 a7aa3b0c47a8cfda50a9c3e9c8b3dd53dbbd296103536b9cd80dc425672e73bc). The author list is the byline as printed on the first page.","versionDate":"2021-12-08","canonicalUrl":"https://arxiv.org/abs/2112.11446v1","evidenceUrl":"https://arxiv.org/pdf/2112.11446v1","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Jack W. Rae and others, “Scaling Language Models: Methods, Analysis & Insights from Training Gopher”, 2021-12-08.","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-xgboost-scalable-tree-boosting-system-2016-v1","documentId":"kp-xgboost-scalable-tree-boosting-system-2016","title":"XGBoost: A Scalable Tree Boosting System","authors":["Tianqi Chen","Carlos Guestrin"],"publisher":"arXiv","publicationDate":"2016-03-09","version":"arXiv:1603.02754v1, submitted 9 March 2016 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 6efbbcf077f2a0344bddb5777a6cd74263a2860e05e55d4be6ca1db911f7c2b5). The author list is the byline as printed on the first page.","versionDate":"2016-03-09","canonicalUrl":"https://arxiv.org/abs/1603.02754v1","evidenceUrl":"https://arxiv.org/pdf/1603.02754v1","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Tianqi Chen, Carlos Guestrin, “XGBoost: A Scalable Tree Boosting System”, 2016-03-09.","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-th-c-zhao-v1","documentId":"th-zhao-2023-llm-survey","title":"A Survey of Large Language Models","authors":["Wayne Xin Zhao","Kun Zhou","Junyi Li","Tianyi Tang"],"publisher":"arXiv","publicationDate":"2023-03-31","version":"Retrieved for the candidate pack on 2026-10-29 (SHA-256 d58be5fc39608dc9aec45194c602436d793f2f2bf56f267d0e38d6258a9f7a9e). arXiv:2303.18223v1 PDF; later revisions (many) not substituted Only the first-listed authors are recorded here; the source credits further authors.","canonicalUrl":"https://arxiv.org/pdf/2303.18223v1","evidenceUrl":"https://arxiv.org/pdf/2303.18223v1","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"Wayne Xin Zhao, Kun Zhou, Junyi Li, Tianyi Tang, “A Survey of Large Language Models”, 2023-03-31.","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-sensetime-prospectus-2021","title":"SenseTime Group Inc. — Global Offering","authors":["SenseTime Group Inc."],"publisher":"SenseTime Group Inc. (via HKEXnews)","publicationDate":"2021-12-07","version":"Dated document, retrieved on 2026-10-29 (SHA-256 45a8fc07d0a356dea0d69b8511e55dd16d4748761910203e5671cf50257195bf) directly from the publisher. Version: HKEXnews 2021120700017.pdf. English prospectus filed on HKEXnews on 7 December 2021 (761 pages). A supplemental prospectus of 20 December 2021 (2021122000009.pdf) was not inspected. Chinese characters in the PDF text layer were not extracted (shown as empty parentheses).","canonicalUrl":"https://www1.hkexnews.hk/listedco/listconews/sehk/2021/1207/2021120700017.pdf","evidenceUrl":"https://www1.hkexnews.hk/listedco/listconews/sehk/2021/1207/2021120700017.pdf","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"SenseTime Group Inc., “SenseTime Group Inc. — Global Offering”, 2021-12-07.","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":"c3-src-step1x-edit-pdf-v1","documentId":"step1x-edit-report","title":"Step1X-Edit: A Practical Framework for General Image Editing (arXiv PDF, v1)","authors":["Step1X-Image Team, StepFun"],"publisher":"arXiv","publicationDate":"2025-04-24","version":"Dated document, retrieved on 2026-10-29 (SHA-256 4566ea2eac28311203cc8d3e28e6bf297d1ab967446cca520a0641f3bbb7f2fe) directly from the publisher. Version: 2504.17761v1. PDF of version 1; later versions not inspected.","canonicalUrl":"https://arxiv.org/pdf/2504.17761v1","evidenceUrl":"https://arxiv.org/pdf/2504.17761v1","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"Step1X-Image Team, StepFun, “Step1X-Edit: A Practical Framework for General Image Editing (arXiv PDF, v1)”, 2025-04-24.","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":"c3-src-step1x-edit-abs-v1","documentId":"step1x-edit-report","title":"Step1X-Edit: A Practical Framework for General Image Editing (arXiv abstract page, v1)","authors":["Step1X-Image Team, StepFun"],"publisher":"arXiv","publicationDate":"2025-04-24","version":"Dated document, retrieved on 2026-10-29 (SHA-256 e1cff26e3e96f862611fae57dacb466980ed2a90d6b632ada212509a0cb0a84c) directly from the publisher. Version: 2504.17761v1. Abstract page of version 1 (submitted 2025-04-24); the page lists later versions v2 to v5 (28 April to 31 July 2025), not inspected.","canonicalUrl":"https://arxiv.org/abs/2504.17761v1","evidenceUrl":"https://arxiv.org/abs/2504.17761v1","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"Step1X-Image Team, StepFun, “Step1X-Edit: A Practical Framework for General Image Editing (arXiv abstract page, v1)”, 2025-04-24.","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":"c3-src-cac-genai-filings-2025-announcement","documentId":"cn-cac-genai-filings-announcement-2026","title":"国家互联网信息办公室关于发布2025年生成式人工智能服务已备案信息的公告","authors":["国家互联网信息办公室"],"publisher":"中央网络安全和信息化委员会办公室 / 国家互联网信息办公室 (cac.gov.cn)","publicationDate":"2026-01-09","version":"Dated page, retrieved on 2026-10-29 (SHA-256 5c9ae18383aad4003699891babcbf58051ba5f383d63db7baac999bffdc68c19) directly from the publisher. CAC page dated 2026年01月09日 18:30, read as it stood on 2026-10-29 (the first two connections were reset; the third succeeded). Its wording can change; earlier wording is not authenticated. The cross-family check (2026-10-31) located the Internet Archive capture of 2026-01-09 linked here; a capture shows only what the archive retrieved that day. Capture of the publication day located by the cross-family check. English title (machine translation): “Announcement of the Cyberspace Administration of China publishing 2025 information on filed generative AI services”.","canonicalUrl":"https://www.cac.gov.cn/2026-01/09/c_1769688009588554.htm","evidenceUrl":"https://www.cac.gov.cn/2026-01/09/c_1769688009588554.htm","language":"zh-Hans","archivedUrl":"https://web.archive.org/web/20260109142356/https://www.cac.gov.cn/2026-01/09/c_1769688009588554.htm","archivedAt":"2026-01-09","accessedAt":"2026-10-29","reuse":{"attribution":"国家互联网信息办公室, “国家互联网信息办公室关于发布2025年生成式人工智能服务已备案信息的公告”, 2026-01-09.","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":"c3-src-cac-genai-list-2026-08","documentId":"cn-cac-genai-service-list-2026-08","title":"生成式人工智能服务已备案和已登记信息（2026年8月）","authors":["国家互联网信息办公室"],"publisher":"国家互联网信息办公室 (cac.gov.cn, attachment to the announcement page of 2 April 2024)","publicationDate":"2026-08","version":"Dated document, retrieved on 2026-10-29 (SHA-256 081f8abad11d3e89cc94e9c4e8f76133613a4f4e05b14d6a88e086be050629e5) directly from the publisher. Version: 附件13: 生成式人工智能服务已备案和已登记信息（2026年8月） (PDF served by /cms/pub/interact/downloadfile.jsp). PDF attachment 13 of the CAC page, retrieved 2026-10-29 through the downloadfile.jsp link on the page (the URL carries an encoded file path and the attachment title). The PDF metadata creation date is 2026-09-14; that is a file property, not a publication date. The month is the one in the attachment title. English title (machine translation): “Information on filed and registered generative AI services (August 2026)”.","canonicalUrl":"https://www.cac.gov.cn/cms/pub/interact/downloadfile.jsp?filepath=NUtqEIwGiCjGm2Bhl20cvGmT22KDqGR7FvHLiywcKRhvz0vHSSJ6sATU6DtmcGo1jaT2QSSyYLABPnQL~026zVAfqphfFn00jUfFRLdBE5I=&fText=%E7%94%9F%E6%88%90%E5%BC%8F%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E6%9C%8D%E5%8A%A1%E5%B7%B2%E5%A4%87%E6%A1%88%E5%92%8C%E5%B7%B2%E7%99%BB%E8%AE%B0%E4%BF%A1%E6%81%AF%EF%BC%882026%E5%B9%B48%E6%9C%88%EF%BC%89","evidenceUrl":"https://www.cac.gov.cn/cms/pub/interact/downloadfile.jsp?filepath=NUtqEIwGiCjGm2Bhl20cvGmT22KDqGR7FvHLiywcKRhvz0vHSSJ6sATU6DtmcGo1jaT2QSSyYLABPnQL~026zVAfqphfFn00jUfFRLdBE5I=&fText=%E7%94%9F%E6%88%90%E5%BC%8F%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E6%9C%8D%E5%8A%A1%E5%B7%B2%E5%A4%87%E6%A1%88%E5%92%8C%E5%B7%B2%E7%99%BB%E8%AE%B0%E4%BF%A1%E6%81%AF%EF%BC%882026%E5%B9%B48%E6%9C%88%EF%BC%89","language":"zh-Hans","accessedAt":"2026-10-29","reuse":{"attribution":"国家互联网信息办公室, “生成式人工智能服务已备案和已登记信息（2026年8月）”, 2026-08.","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-narayanan-kapoor-2024-ai-scaling-myths","documentId":"narayanan-kapoor-2024-ai-scaling-myths","title":"AI scaling myths","authors":["Arvind Narayanan","Sayash Kapoor"],"publisher":"AI Snake Oil (the authors' Substack, now at normaltech.ai)","publicationDate":"2024-06-27","version":"Retrieved for the candidate pack on 2026-10-29 (SHA-256 14de787679324dbfbfdaaf1093d0c62ecba3e07a55732cdfa9ab74cc467f913d). Current Substack page (requested at aisnakeoil.com, served at normaltech.ai), dated 2024-06-27 in its metadata; retrieved 2026-10-29.","canonicalUrl":"https://www.normaltech.ai/p/ai-scaling-myths","evidenceUrl":"https://www.normaltech.ai/p/ai-scaling-myths","language":"en","accessedAt":"2026-10-29","reuse":{"attribution":"Arvind Narayanan, Sayash Kapoor, “AI scaling myths”, 2024-06-27.","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-prov-fsf-v2-0-hold-pending-mitigations","claimId":"prov-fsf-v2-0-hold-pending-mitigations","sourceId":"source-fsf-2-0","sourceRole":"primary","support":"supports","locator":"PDF p. 3 (printed p. 3), Framework, component “3 - Applying Mitigations”, third paragraph, second sentence","passage":"If this happens, the response plan may involve putting deployment or further development on hold until adequate mitigations can be applied.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the Internet Archive capture of 2025-02-05 (the edition as first published) and the current file, which carries a later documented correction; the passage is identical in both. Extraction note: “If this happens” refers to the paragraph’s first sentence, restored in the statement; the two sentences together are 51 words. Same wording and page in both copies. Human editorial review is pending."},{"id":"evidence-absent-rsp-v3-4-no-train-deploy-without-safeguards","claimId":"absent-rsp-v3-4-no-train-deploy-without-safeguards","sourceId":"source-rsp-3-4","sourceRole":"primary","support":"supports","locator":"RSP v3.4 (effective July 8, 2026), whole text layer, PDF pp. 1–21: Introduction (pp. 3–4); §1 with the threshold table (pp. 4–10, columns on p. 9 read individually because the text layer interleaves them); §2–§3 Risk Reports (pp. 11–16); §4 Governance (p. 16); Appendix A and Appendix B (p. 17); Changelog (pp. 18–21).","passage":"Within the inspected scope, no passage stating the provision “Commitment not to train or deploy without adequate safeguards” was found.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-03","note":"Absence recorded by an automated research assistant on 2026-10-03. Method: Same reading, comparison and case-insensitive search terms as for v3.0, plus a word-level comparison with v3.3 (differences: Automated R&D threshold, Risk Report coverage date, internal sharing, redaction disclosure, external review split). No passage stating a commitment not to train or deploy models unless adequate safeguards are implemented, or a commitment to pause training or deployment until such safeguards are in place, was found within the operative text. Hits examined and not counted: the Introduction (p. 3), §1 (p. 4) and Appendix A scenario commitments (p. 17) as in v3.0; the Appendix A introduction sentences on cautionary action and considering pausing (p. 17), as in v3.1; the reproduced RSP v2.0 changelog entry (p. 18). Human editorial review is pending."},{"id":"evidence-sb-attorney-general-enforcement-1","claimId":"sb-attorney-general-enforcement","sourceId":"source-ca-sb53-chaptered","sourceRole":"primary","support":"supports","locator":"SEC. 2, Bus. & Prof. Code § 22757.15(b)","passage":"A civil penalty described in this section shall be recovered in a civil action brought only by the Attorney General.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-04","note":"Verbatim excerpt proposed by a research agent and checked by two independent fresh-context verifier agents and decided by a third adjudicating agent on 2026-10-04; re-matched mechanically by the integrator against a direct fetch of the cited expression. Matched against the live bill page, the capture and the live codified section. Human editorial review is pending."},{"id":"evidence-sb-attorney-general-enforcement-2","claimId":"sb-attorney-general-enforcement","sourceId":"source-ca-bpc-22757-15","sourceRole":"primary","support":"context","locator":"History note at the end of § 22757.15","passage":"the section's history note records that it was added by Stats. 2025, ch. 138, sec. 2 (SB 53), effective January 1, 2026. It lists no amendment.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-04","note":"Editorial paraphrase proposed by a research agent and checked by two independent fresh-context verifier agents and decided by a third adjudicating agent on 2026-10-04. Paraphrased because the codified page is only about 116 words. Supports the effective date in the time note. The CDX API lists the capture with HTTP 200, but I could not open its content because the Wayback Machine returned a bot-block page. Context: the codified section and its history note as displayed when checked; the provision itself is quoted from the chaptered text. Human editorial review is pending."},{"id":"evidence-cs-openai-group-pbc-legal-form-2025-1","claimId":"cs-openai-group-pbc-legal-form-2025","sourceId":"source-openai-group-pbc-certificate","sourceRole":"primary","support":"supports","locator":"Page 1, filing stamp and title","passage":"The stamp reads State of Delaware, Secretary of State, Division of Corporations, Delivered 08:06 AM 10/28/2025, FILED 08:07 AM 10/28/2025, SR 20254387187, File Number 10381551; the title reads Certificate of Incorporation of OpenAI Group PBC, Delaware Public Benefit Corporation.","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. Added by the integrating session in week 4c; not part of the week 4b research pass. Human editorial review is pending."},{"id":"evidence-cs-openai-group-pbc-legal-form-2025-2","claimId":"cs-openai-group-pbc-legal-form-2025","sourceId":"source-openai-group-pbc-certificate","sourceRole":"primary","support":"supports","locator":"Article III, first paragraph (page 1)","passage":"The Corporation shall be a public benefit corporation as contemplated by Subchapter XV of the DGCL","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. Added by the integrating session in week 4c; not part of the week 4b research pass. Human editorial review is pending."},{"id":"evidence-ts-a3-iem","claimId":"ts-a3-iem","sourceId":"source-iem","sourceRole":"primary","support":"supports","locator":"Abstract, second paragraph","passage":"I identify the key issue as returns on cognitive reinvestment—the ability to invest more computing power, faster computers, or improved cognitive algorithms to yield cognitive labor which produces larger brains, faster brains, or better mind designs.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"The same passage as the A1 record."},{"id":"evidence-ts-b3-agi-ruin","claimId":"ts-b3-agi-ruin","sourceId":"source-agi-ruin-2022","sourceRole":"primary","support":"supports","locator":"Section B.1 \"The distributional leap\", item 13","passage":"'change your outer behavior to deliberately look more aligned and deceive the programmers, operators, and possibly any loss functions optimizing over you'. This problem is one that will appear at the superintelligent level","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Checked in the first stored version on 2026-10-06. On 2026-10-30, by the maintainer’s decision of that day, the quotation was corrected to the archive capture of 2022-06-05 (quotation marks and apostrophes as in the capture), which a cross-family archive check read on 2026-10-28; the proposition is unchanged."},{"id":"evidence-ec-g06-d","claimId":"ec-g06-clancy-besiroglu-2023-great-inflection-a","sourceId":"source-clancy-besiroglu-2023-great-inflection","sourceRole":"primary","support":"supports","locator":"Matt Clancy's turn beginning \"As a quick aside\"","passage":"note that the 100-million-fold increase in computing power dedicated to AI over the last decade has not led to an acceleration in economic growth so far.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage G06-d (research/economics-growth). The quotation (26 words) matched the checked expression (asterisk-debate) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day, whose SHA-256 differs from the recorded one."},{"id":"evidence-ec-m05-20","claimId":"ec-m05-acemoglu-restrepo-2016-race-between-machine-and-man-a","sourceId":"source-acemoglu-restrepo-2016-race-between-machine-and-man","sourceRole":"primary","support":"supports","locator":"pdf page 38; section 6 Conclusion (printed p. 36)","passage":"If in the long run capital is very cheap relative to labor, automation technologies will advance rapidly and labor will become redundant.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M05-20 (research/economics-mainstream). The quotation (22 words) matched the retrieved current copy (SHA-256 441d01202afd56ef8002fc24ffc2beb51191741c0b5accb11d2534620dd616b7) 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":"Conditional on a very low long-run rental rate of capital relative to wages. Otherwise there is a stable balanced growth path with new tasks for labour, which the authors treat as the case under reasonable assumptions."}},{"id":"evidence-ec-m06b-18","claimId":"ec-m06b-agrawal-gans-goldfarb-2018-economic-policy-for-ai-a","sourceId":"source-agrawal-gans-goldfarb-2018-economic-policy-for-ai","sourceRole":"primary","support":"supports","locator":"pdf page 23; section 4 Conclusions","passage":"Addressing these consequences will be the role of education policy, the social safety net, and antitrust enforcement.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M06b-18 (research/economics-mainstream). The quotation (17 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":"evidence-ec-m09-11","claimId":"ec-m09-nordhaus-2015-economic-singularity-b","sourceId":"source-nordhaus-2015-nber","sourceRole":"primary","support":"supports","locator":"pdf page 10; Section III, Historical perspectives on Singularity in economics (printed p. 7)","passage":"There is remarkably little writing on Singularity in the modern macroeconomic literature.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M09-11 (research/economics-mainstream). The quotation (12 words) matched the retrieved current copy (SHA-256 cb481e5e99ef9bad8575d9150bcc8464eda4904722cc5048389ab6159858b062) 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 immediately preceding paragraphs discuss Herbert Simon's 'The Shape of Automation' (1965) as a 'pathbreaking' macro analysis of computerisation, plus Keynes 1930 and endogenous growth. These are counter-evidence to a general 'only recently' reading."}},{"id":"evidence-ec-m09-12","claimId":"ec-m09-nordhaus-2015-economic-singularity-b","sourceId":"source-nordhaus-2015-nber","sourceRole":"primary","support":"supports","locator":"pdf page 5; Section II, Artificial Intelligence and the Singularity (printed p. 2)","passage":"For those with a background primarily in economics, the present section is likely to read like science fiction.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M09-12 (research/economics-mainstream). The quotation (18 words) matched the retrieved current copy (SHA-256 cb481e5e99ef9bad8575d9150bcc8464eda4904722cc5048389ab6159858b062) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m09-35","claimId":"ec-m09-nordhaus-2015-economic-singularity-b","sourceId":"source-nordhaus-2015-nber","sourceRole":"primary","support":"supports","locator":"pdf page 9; section on the history of the accelerationist debate","passage":"One of the most impressive attempts to deal with the macroeconomic implications of computerization was Herbert Simon in \"The Shape of Automation\" (1965).","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M09-35 (research/economics-mainstream). The quotation (23 words) matched the retrieved current copy (SHA-256 cb481e5e99ef9bad8575d9150bcc8464eda4904722cc5048389ab6159858b062) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m09-33","claimId":"ec-m09-calvano-et-al-2020-algorithmic-pricing-collusion-b","sourceId":"source-calvano-et-al-2020-algorithmic-pricing-collusion","sourceRole":"primary","support":"supports","locator":"AEA abstract page; Abstract (AER 110(10), October 2020, pp. 3267–97)","passage":"we study experimentally the behavior of algorithms powered by Artificial Intelligence (Q-learning) in a workhorse oligopoly model of repeated price competition.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M09-33 (research/economics-mainstream). The quotation (21 words) matched the retrieved current copy (SHA-256 48c037a5cf242fca6ad8f4c36d38f34e248b31d856fb6b1d9475eed61f5925be) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day, whose SHA-256 differs from the recorded one."},{"id":"lg-ev-mx-nordeen-xai-team-2023","claimId":"lg-mx-nordeen-xai-team-2023","sourceId":"source-xai-team-2023","sourceRole":"secondary","support":"supports","locator":"Paragraph beginning 'That list includes', second sentence","passage":"Ars Technica lists Ross Nordeen among the team named on xAI's website.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “Ross Nordeen” 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-oai-karpathy-departure-2024","claimId":"lg-oai-karpathy-departure-2024","sourceId":"lg-src-oai-karpathy-x-2024","sourceRole":"primary","support":"supports","locator":"Post text, first sentence","passage":"Karpathy writes that he left OpenAI the day before the post.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “I left OpenAI yesterday” 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-deal-google-windsurf-agreement-licence","claimId":"lg-deal-google-windsurf-agreement-2025","sourceId":"lg-src-deal-verge-windsurf-2025","sourceRole":"secondary","support":"supports","locator":"Paragraph 2","passage":"Google will not have any control over nor a stake in Windsurf, but it will take a non-exclusive license to some of Windsurf’s technology.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (24 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 judged it partly supported; its findings were applied at admission. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-deal-google-windsurf-agreement-hiring","claimId":"lg-deal-google-windsurf-agreement-2025","sourceId":"lg-src-deal-verge-windsurf-2025","sourceRole":"secondary","support":"supports","locator":"Paragraph 1","passage":"Google will instead hire Windsurf CEO Varun Mohan, cofounder Douglas Chen, and some of Windsurf’s R&D employees","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (17 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 judged it partly supported; its findings were applied at admission. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-deal-google-windsurf-agreement-windsurf","claimId":"lg-deal-google-windsurf-agreement-2025","sourceId":"lg-src-deal-windsurf-next-stage-2025","sourceRole":"primary","support":"supports","locator":"Paragraph beginning “We’re excited to announce” (the third body paragraph of the current copy)","passage":"As part of the agreement, Varun, Douglas, and some members of our R&D team will join Google.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (17 words) 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. A cross-family machine check (2026-10-16/17) re-read the record’s sources and judged it partly supported; its findings were applied at admission. No human editorial review.","retrievalRoute":"reader_relay"},{"id":"lg-ev-ant-cuellar-ltbt-trustee-1","claimId":"lg-ant-cuellar-ltbt-trustee","sourceId":"source-anthropic-ltbt-2023","sourceRole":"primary","support":"supports","locator":"Footnote on changes in the Trust's composition (live page)","passage":"Anthropic’s Trust page states that Mariano-Florentino (Tino) Cuéllar, President of the Carnegie Endowment for International Peace, joined in January 2026.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “joined in January 2026” 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. Shortened at admission to keep the quotations from this source within the per-document limit; the shortened wording is part of the checked passage. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-ant-cuellar-ltbt-trustee-2","claimId":"lg-ant-cuellar-ltbt-trustee","sourceId":"lg-src-ant-cuellar-2026","sourceRole":"primary","support":"supports","locator":"Paragraph beginning “Tino has served as a Trustee”","passage":"Tino has served as a Trustee of Anthropic's Long-Term Benefit Trust since January 2026.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (14 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. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-ant-cuellar-ltbt-trustee-company-page","claimId":"lg-ant-cuellar-ltbt-trustee","sourceId":"lg-src-ant-company-2026","sourceRole":"primary","support":"context","locator":"Trustee list (page retrieved on 15 October 2026)","passage":"The Trustee list on the page retrieved on 15 October 2026 names Neil Buddy Shah, Richard Fontaine and Ben Bernanke.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “Neil Buddy Shah, Richard Fontaine, and Dr. Ben Bernanke” 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. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-oai-kc-cpo-2026","claimId":"lg-oai-kc-cpo-2026","sourceId":"lg-src-oai-kc-2026","sourceRole":"primary","support":"supports","locator":"Opening sentence","passage":"welcome Arvind KC to OpenAI as our Chief People Officer","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (10 words) 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. No human editorial review.","retrievalRoute":"reader_relay"},{"id":"lg-ev-cn-junyang-lin-qwen-departure-2026","claimId":"lg-cn-junyang-lin-qwen-departure-2026","sourceId":"lg-src-cn-lin-junyang-x-2026","sourceRole":"primary","support":"supports","locator":"X post by @JustinLin610","passage":"Lin wrote that he was stepping down and said goodbye to Qwen.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “stepping down” 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-cn-junyang-lin-qwen-departure-2026-caixin","claimId":"lg-cn-junyang-lin-qwen-departure-2026","sourceId":"lg-src-cn-caixin-lin-2026","sourceRole":"secondary","support":"supports","locator":"Opening paragraph","passage":"Lin Junyang, a technical leader of Alibaba Group Holding Ltd.’s Qwen AI model team, has resigned","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (16 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. No human editorial review.","retrievalRoute":"direct"},{"id":"ev-cmp-fsf-v2-0-d6-discretion-1","claimId":"cmp-fsf-v2-0-d6-discretion","sourceId":"source-fsf-2-0","sourceRole":"primary","support":"supports","locator":"PDF p. 5 (printed p. 5), Misuse Critical Capability Levels, paragraph beginning 'These recommended security levels', second sentence, final clause","passage":"or if we assess that the benefits of the open release of model weights outweigh the risks.","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. A second example of when recommended security levels may be adjusted. Human editorial review is pending."},{"id":"evidence-ap-eliezer-yudkowsky-sl4-2000-08-21-yudkowsky-0008-0046","claimId":"ap-eliezer-yudkowsky-sl4-2000-08-21-yudkowsky-0008-0046","sourceId":"source-sl4-2000-08-21-yudkowsky-0008-0046","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-lw-2022-six-dimensions-of-operational-adequacy-in","claimId":"ap-eliezer-yudkowsky-lw-2022-six-dimensions-of-operational-adequacy-in","sourceId":"source-lw-2022-six-dimensions-of-operational-adequacy-in","sourceRole":"primary","support":"supports","locator":"Post author list in the publisher’s post data (LessWrong GraphQL API)","passage":"Eliezer 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 through the publisher’s post data. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"ev-cmp-rsp-v2-0-d5-cadence-1","claimId":"cmp-rsp-v2-0-d5-cadence","sourceId":"source-rsp-2-0","sourceRole":"primary","support":"supports","locator":"PDF p. 9 (printed p. 5), §3.1 Preliminary Assessment, definition of “notably more capable” and numbered items 1–2","passage":"The edition treats a model as “notably more capable”, which triggers a comprehensive assessment, when it scores notably higher in automated testing of risk-relevant domains (an Effective Compute increase of at least 4x) or six months of finetuning and elicitation improvements have accumulated.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-28","note":"Explicit paraphrase by an automated research assistant, corrected on 2026-10-28 after a blind check of all 287 comparison paraphrases by a reviewer from a different model family on 2026-10-28 (research/frontier-framework-comparison/codex-paraphrases-2026-10-28.md), which found that its locator named only item 1, while the six-month alternative is item 2; the locator was corrected and the paraphrase is unchanged. The correction was written from the passage as re-read in the publisher copy retrieved on 2026-10-28, whose SHA-256 matches the copy retrieved on 2026-10-15. Given as a paraphrase to keep the edition within the quotation limit. The Responsible Scaling Officer may also require a comprehensive assessment at their discretion. Human editorial review is pending."},{"id":"ev-cmp-rsp-v2-1-d1-other-1","claimId":"cmp-rsp-v2-1-d1-other","sourceId":"source-rsp-2-1","sourceRole":"primary","support":"supports","locator":"PDF p. 6 (printed p. 2), §1 Background, paragraph after the Deployment/Security Standards bullets","passage":"The edition says it expects to refine the framework further as future risks arise, giving as an example the risk of an AI system trying to subvert its operators’ goals.","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. Human editorial review is pending."},{"id":"evidence-fl-eu-aia-annex13-criteria-1","claimId":"fl-eu-aia-annex13-criteria","sourceId":"source-eu-ai-act-2024","sourceRole":"primary","support":"supports","locator":"Annex XIII, points (a) to (g)","passage":"The criteria are (a) number of parameters; (b) quality or size of the data set; (c) training computation or proxies such as cost, time or energy; (d) input and output modalities and per-modality thresholds; (e) benchmarks and evaluations including autonomy, scalability and tool access; (f) high impact on the internal market due to reach; (g) number of registered end-users.","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-eu-aia-annex13-criteria-2","claimId":"fl-eu-aia-annex13-criteria","sourceId":"source-eu-ai-act-2024","sourceRole":"primary","support":"supports","locator":"Annex XIII, point (f)","passage":"which shall be presumed when it has been made available to at least 10 000 registered business users established in the Union","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (22 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-cpo-ch-nvidia-openai-investment-intent-1","claimId":"cpo-ch-nvidia-openai-investment-intent","sourceId":"source-cpo-ch-openai-nvidia-2025","sourceRole":"primary","support":"supports","locator":"highlights, bullet 2","passage":"NVIDIA intends to invest up to $100 billion in OpenAI progressively as each gigawatt is deployed.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (16 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 (SHA-256 ffa6da3512fd4bc20607cc222d5a9cabc868805dc17fb6764d23601774e8d275; the bytes differ from the pack’s retrieval). 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 66) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-orb-spacex-odc-plan-1","claimId":"cpo-orb-spacex-odc-plan","sourceId":"source-cpo-orb-spacex-xai","sourceRole":"primary","support":"supports","locator":"paragraph 5","passage":"Launching a constellation of a million satellites that operate as orbital data centers","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 112) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-orb-spacex-odc-plan-2","claimId":"cpo-orb-spacex-odc-plan","sourceId":"source-cpo-orb-spacex-xai","sourceRole":"primary","support":"context","locator":"paragraph 9 (\"The basic math\")","passage":"the letter computes that launching a million tons of satellites a year at 100 kW of compute per ton would add 100 GW of AI compute capacity a year.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The paraphrase is Selekoa’s, written from the text retrieved for the candidate pack on 2026-10-28; it is not a quotation and was not machine-matched. The cross-family check found it faithful. A cross-family machine check (2026-10-29/30, item 112) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-kp-states-gpt-4-technical-report-2023-1-1","claimId":"kp-states-gpt-4-technical-report-2023-1","sourceId":"source-kp-gpt-4-technical-report-2023-v1","sourceRole":"primary","support":"supports","locator":"arXiv v1, PDF p. 1, abstract","passage":"We report the development of GPT-4, a large-scale, multimodal model which can accept image and text inputs and produce text outputs.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The quotation (21 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. No human editorial review."},{"id":"evidence-kp-states-constitutional-ai-harmlessness-from-ai-2022-1-1","claimId":"kp-states-constitutional-ai-harmlessness-from-ai-2022-1","sourceId":"source-kp-constitutional-ai-harmlessness-from-ai-2022-v1","sourceRole":"primary","support":"supports","locator":"arXiv v1, PDF p. 2, §1 Introduction, para. 2","passage":"In this paper we develop a method we refer to as Constitutional AI (CAI), depicted in Figure 1, and use it to train a non-evasive and relatively harmless AI assistant, without any human feedback labels for harms.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The quotation (37 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. No human editorial review."},{"id":"evidence-kp-states-xgboost-scalable-tree-boosting-system-2016-1-1","claimId":"kp-states-xgboost-scalable-tree-boosting-system-2016-1","sourceId":"source-kp-xgboost-scalable-tree-boosting-system-2016-v1","sourceRole":"primary","support":"supports","locator":"arXiv v1, PDF p. 1, §1 Introduction, para. 3","passage":"In this paper, we describe XGBoost, a scalable machine learning system for tree boosting.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, key papers (research/key-papers). The quotation (14 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 B031) found the quotation at its locator. No human editorial review."},{"id":"evidence-kp-cites-iem-2013-basic-ai-drives-2008-1","claimId":"kp-cites-iem-2013-basic-ai-drives-2008","sourceId":"source-iem","sourceRole":"primary","support":"supports","locator":"PDF p. 94, 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 with author and year) 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-kimi-k2-report-triviaqa-large-scale-distantly-supervised-2017-1","claimId":"kp-cites-kimi-k2-report-triviaqa-large-scale-distantly-supervised-2017","sourceId":"c2-src-kimi-k2-pdf-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 22, reference by arXiv identifier","passage":"The text gives the arXiv identifier 1705.03551.","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-chatglm-report-llama-open-efficient-foundation-language-2023-1","claimId":"kp-cites-chatglm-report-llama-open-efficient-foundation-language-2023","sourceId":"source-cn-chatglm","sourceRole":"primary","support":"supports","locator":"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. A cross-family machine check (2026-10-30, item B081) found the citation in the citing text; it qualified its answer, and the adjudication of 2026-10-31 was applied at admission. No human editorial review."},{"id":"evidence-kp-cites-self-critique-paper-self-consistency-improves-chain-thought-2022-1","claimId":"kp-cites-self-critique-paper-self-consistency-improves-chain-thought-2022","sourceId":"source-self-critique-paper-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 23, reference by arXiv identifier","passage":"The text gives the arXiv identifier 2203.11171.","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. A cross-family machine check (2026-10-30, item B091) found the citation in the citing text. No human editorial review."},{"id":"evidence-kp-authored-niki-parmar-attention-all-you-need-1","claimId":"kp-authored-niki-parmar-attention-all-you-need","sourceId":"source-attention-all-you-need-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 1, byline","passage":"Niki Parmar","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. A cross-family machine check (2026-10-30, item A090) transcribed the byline blind. No human editorial review."},{"id":"evidence-kp-authored-alec-radford-ppo-paper-1","claimId":"kp-authored-alec-radford-ppo-paper","sourceId":"source-ppo-paper-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 1, byline","passage":"Alec Radford","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. A cross-family machine check (2026-10-30, item A079) transcribed the byline blind. No human editorial review."},{"id":"evidence-kp-authored-laurent-sifre-scaling-language-models-methods-analysis-2021-1","claimId":"kp-authored-laurent-sifre-scaling-language-models-methods-analysis-2021","sourceId":"source-kp-scaling-language-models-methods-analysis-2021-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 1, 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-th-term-llm-zhao-2023-def-1","claimId":"th-term-llm-zhao-2023-def","sourceId":"source-th-c-zhao-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 3, section 2, \"Background\"","passage":"Typically, large language models (LLMs) refer to language models that contain hundreds of billions (or more) of parameters","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, Kinds of AI and technical history (research/technical-history). 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. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-th-term-llm-zhao-2023-def-2","claimId":"th-term-llm-zhao-2023-def","sourceId":"source-th-c-zhao-v1","sourceRole":"primary","support":"supports","locator":"PDF p. 3, section 2, \"Background\"","passage":"Specifically, LLMs are built upon the Transformer architecture","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, Kinds of AI and technical history (research/technical-history). The quotation (8 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-llm-zhao-2023-def-3","claimId":"th-term-llm-zhao-2023-def","sourceId":"source-th-c-zhao-v1","sourceRole":"primary","support":"context","locator":"PDF p. 31, section 8","passage":"We focus on the discussion of large-sized models (i.e., having a size larger than 10B)","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":"c3-ev-sensetime-xu-li-chair-ceo-2021","claimId":"c3-sensetime-xu-li-chair-ceo-2021","sourceId":"c3-src-sensetime-prospectus-2021","sourceRole":"primary","support":"supports","locator":"Directors and Senior Management, Executive Directors","passage":"is our co-founder, executive Chairman of our Board, executive Director and chief executive officer. He was appointed as a Director on December 15, 2015 and re-designated as an executive Director on August 23, 2021.","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 (34 words) 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. A cross-family machine check (2026-10-31, item K026) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"c3-ev-stepfun-step1x-edit-report","claimId":"c3-stepfun-step1x-edit-report","sourceId":"c3-src-step1x-edit-pdf-v1","sourceRole":"primary","support":"supports","locator":"Title page","passage":"Step1X-Image Team StepFun","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 (3 words) 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. No human editorial review.","retrievalRoute":"direct"},{"id":"c3-ev-stepfun-step1x-edit-report-2","claimId":"c3-stepfun-step1x-edit-report","sourceId":"c3-src-step1x-edit-abs-v1","sourceRole":"primary","support":"supports","locator":"arXiv abstract page, abstract","passage":"we aim to release a state-of-the-art image editing model, called Step1X-Edit","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 (11 words) 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. No human editorial review.","retrievalRoute":"direct"},{"id":"c3-ev-governance-cac-genai-list-2026-08-count-registered-1","claimId":"c3-governance-cac-genai-list-2026-08-count-registered","sourceId":"c3-src-cac-genai-list-2026-08","sourceRole":"primary","support":"supports","locator":"Last registered row (p. 12)","passage":"Table rows counted: 133 (registered (已登记) applications or functions added in this list, headed 生成式人工智能服务已登记信息; rows numbered 599 to 731 (cumulative number reached: 731); counted by the preparer).","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, Chinese frontier-AI institutions, third collection (research/china-labs-3). The paraphrase’s verbatim anchor 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. The paraphrase is Selekoa’s; only the anchor is the source’s wording. English machine translation, not verified by a person: “Row 731: Tianjin; job-seeking and recruitment intelligent adviser”. A cross-family machine check (2026-10-31, item K113) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"c3-ev-governance-cac-genai-list-2026-08-count-registered-registration-announcement","claimId":"c3-governance-cac-genai-list-2026-08-count-registered","sourceId":"c3-src-cac-genai-filings-2025-announcement","sourceRole":"primary","support":"context","locator":"Body, first paragraph","passage":"Applications or functions that call filed models through an API or otherwise are registered by local cyberspace offices; 330 were newly registered in 2025, and 435 had completed registration cumulatively as of 31 December 2025.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-31","note":"Candidate pack passage, Chinese frontier-AI institutions, third collection (research/china-labs-3). The paraphrase’s verbatim anchor 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. The paraphrase is Selekoa’s; only the anchor is the source’s wording. English machine translation, not verified by a person: “Generative AI applications or functions that directly call the capabilities of filed models through an API or other means are registered by local cyberspace offices; 330 new ones completed registration … 435 generative AI applications or functions had completed registration”. Cited at the admission of the third Chinese collection (research/china-labs-3) as context for the January 2026 announcement’s description of registration, which this record’s qualification gives; the same passage supports another record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-rd-reply-narayanan-kapoor-2024-ai-scaling-myths-to-aschenbrenner-2024-situational-awareness-1","claimId":"rd-reply-narayanan-kapoor-2024-ai-scaling-myths-to-aschenbrenner-2024-situational-awareness","sourceId":"source-rd-narayanan-kapoor-2024-ai-scaling-myths","sourceRole":"primary","support":"supports","locator":"Section \"Further reading\"","passage":"A recent essay by Leopold Aschenbrenner made waves due to its claim that “AGI by 2027 is strikingly plausible”. We haven’t tried to give a point-by-point rebuttal here","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 (28 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-reply-narayanan-kapoor-2024-ai-scaling-myths-to-aschenbrenner-2024-situational-awareness-2","claimId":"rd-reply-narayanan-kapoor-2024-ai-scaling-myths-to-aschenbrenner-2024-situational-awareness","sourceId":"source-rd-narayanan-kapoor-2024-ai-scaling-myths","sourceRole":"primary","support":"supports","locator":"Section \"Further reading\"","passage":"His arguments for his timeline are entertaining and thought provoking, but fundamentally an exercise in trendline extrapolation.","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"}],"at":{"entities":[0,3,7,19,21,22,23,28,39,42,43,44,45,46,47,48,49,50,51,52,71,87,91,102,149,165,180,181,194,205,212,214,333,338,434,447,491,509,517,519,522,526,532,538,546,562,633,724,738,754,782,793,803,822,964,1168,1457,1458,1468,1478,1502,1503,1508,1539,1647,1667,1683,1729,1765,1784,1884],"claims":[435,489,617,643,935,971,1281,1370,1430,1457,1472,1659,1674,1789,1829,1851,1853,2107,2380,2466,2500,2509,2863,3213,3234,3776,3790,3846,4108,4275,4364,4409,4461,4544,4663,4877,4941,5019,5086,5359],"sources":[10,22,36,37,43,45,87,91,101,173,191,197,215,226,338,401,414,457,483,489,490,498,515,563,564,579,581,603,604,745,746,792,925,946,1278,1301,1384,1402,1569,1579,1639,1640,1657,1673,1783],"evidence":[467,520,718,719,775,776,1157,1195,1603,1713,1780,1815,1816,1817,1838,2089,2110,2298,2299,2300,2349,2350,2351,2385,2388,2389,2730,3027,3113,3147,3156,3666,3667,4374,4434,4435,5234,5248,5304,5566,5733,5822,5867,5919,6002,6121,6387,6388,6389,6491,6624,6625,6729,6730,7199,7200]},"version":"131evrz"}