{"entities":[{"id":"openai","name":"OpenAI"},{"id":"xai","name":"xAI"},{"id":"ilya-sutskever","name":"Ilya Sutskever"},{"id":"eliezer-yudkowsky","name":"Eliezer Yudkowsky"},{"id":"deceptive-alignment","name":"Deceptive alignment"},{"id":"rsp-v2-1","name":"Anthropic Responsible Scaling Policy — v2.1"},{"id":"fsf-v2-0","name":"Google DeepMind Frontier Safety Framework — v2.0"},{"id":"christian-szegedy","name":"Christian Szegedy"},{"id":"meta","name":"Meta"},{"id":"joelle-pineau","name":"Joelle Pineau"},{"id":"meta-aasf-v2-0","name":"Meta Advanced AI Scaling Framework — Version 2"},{"id":"ms-fgf-2026","name":"Microsoft Frontier Governance Framework — February 2026"},{"id":"california-sb-53","name":"California SB 53 (2025): Transparency in Frontier Artificial Intelligence Act, as chaptered"},{"id":"openai-inc","name":"OpenAI, Inc. (OpenAI Foundation)"},{"id":"moravec-1989-genetic-takeover","name":"Human Culture—A Genetic Takeover Underway"},{"id":"rosenfelder-1991-cheating-turing-test","name":"Re: Cheating on the Turing Test"},{"id":"vinge-1993-coming-technological-singularity","name":"The Coming Technological Singularity: How to Survive in the Post-Human Era"},{"id":"arbital-programmer-deception","name":"Programmer deception"},{"id":"observed-behaviour-as-evidence","name":"Observed behaviour under training or oversight as evidence"},{"id":"short-instruction-suffices","name":"A short instruction such as making everyone happy"},{"id":"feedback-instils-values","name":"Training on human feedback instils human values"},{"id":"give-it-human-values","name":"Giving it good human values"},{"id":"lw-2008-what-i-think-if-not-why","name":"What I Think, If Not Why"},{"id":"sl4-2000-10-03-yudkowsky-0010-0125","name":"Re: Ben what are your views and concerns"},{"id":"sl4-2002-06-27-yudkowsky-0206-4453","name":"Re: Military Friendly AI"},{"id":"sl4-2005-08-09-yudkowsky-0508-11866","name":"Re: On Our Duty to Not Be Responsible for Artificial Minds"},{"id":"machines-cheaper-at-every-task","name":"Machines cheaper than labour at every task"},{"id":"armer-1966-computer-aspects-of-technological-change","name":"Computer Aspects of Technological Change (selection)"},{"id":"anna-makanju","name":"Anna Makanju"},{"id":"luke-metz","name":"Luke Metz"},{"id":"hans-moravec","name":"Hans Moravec"},{"id":"uk-anthropic-mou-2025","name":"Memorandum of Understanding between the UK and Anthropic on AI opportunities"},{"id":"openai-gpt5-system-card-2025","name":"GPT-5 System Card"}],"claims":[{"id":"sutskever-openai-2015","statement":"OpenAI’s launch announcement names Ilya Sutskever as research director.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"This is a historical role snapshot. Neither the appointment date nor the end of the role is established by this record.","evidenceIds":["evidence-sutskever-openai-2015"],"attributes":{"role":"Research director"}},{"id":"xm-pineau-meta","statement":"The FAIR retrospective identifies Joelle Pineau as Meta VP of AI Research.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A byline role snapshot; no exact appointment date or continuing role is inferred.","evidenceIds":["evidence-xm-pineau-meta"],"attributes":{"role":"VP of AI Research"}},{"id":"prov-fsf-v2-0-pre-deployment-review","statement":"FSF v2.0 states that general availability deployment of a model takes place only after the appropriate corporate governance body determines the safety case regarding each CCL the model has reached to be adequate.","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-pre-deployment-review"],"attributes":{"provision":{"kind":"decision_authority","key":"fsf-pre-deployment-review","parties":[{"role":"decides","actor":"Appropriate corporate governance body"}]}}},{"id":"prov-meta-aasf-v2-0-critical-development-measure","statement":"Advanced AI Scaling Framework v2 states that, should either critical-threshold scenario arise, Meta will only continue development of the Frontier AI if its risk assessments are complete and safeguards are defined, implemented and validated to reduce risk to the moderate or lower risk threshold.","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-meta-aasf-v2-0-critical-development-measure"],"attributes":{"provision":{"kind":"required_action","key":"meta-critical-development-measure"}}},{"id":"prov-rsp-v2-1-policy-change-approval","statement":"RSP v2.1 states that changes to the policy will be proposed by the CEO and the Responsible Scaling Officer and approved by the Board of Directors, in consultation with the Long-Term Benefit Trust.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the edition states, quoted from the cited expression. It is not evidence of implementation, compliance or effectiveness.","evidenceIds":["evidence-prov-rsp-v2-1-policy-change-approval"],"attributes":{"provision":{"kind":"decision_authority","key":"rsp-policy-change-approval","parties":[{"role":"recommends","actor":"CEO"},{"role":"recommends","actor":"Responsible Scaling Officer"},{"role":"decides","actor":"Board of Directors"},{"role":"consulted","actor":"Long-Term Benefit Trust"}]}}},{"id":"sb-link-ms-fgf-2026","statement":"Microsoft's Frontier Governance Framework (February 2026) explicitly references California's TFAIA in two places. Its change log describes the edition as a one-year update following finalisation of the EU GPAI Code of Practice, New York's RAISE Act and TFAIA. Its leading-indicator assessment is run on any model in scope for frontier-model requirements under applicable laws, and it names TFAIA among those laws.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The edition’s own statement about SB 53, quoted from it. It is not evidence that the edition satisfies the statute, that its publisher is a frontier developer under it, or of implementation or compliance. The inspected passages do not say that this document is Microsoft's TFAIA frontier AI framework, or that Microsoft is a large frontier developer; neither is inferred. The change log also lists aligning the update cadence with regulatory obligations and aligning the framework's scope with that of regulations. It does not say which regulation drives which change.","evidenceIds":["evidence-sb-link-ms-fgf-2026-1","evidence-sb-link-ms-fgf-2026-2"],"attributes":{"commitment":{"link":"responds_to"}}},{"id":"ts-b3-rosenfelder-1991-context","statement":"Rosenfelder’s 1991 message proposes conversational tricks a program could use to seem more capable to a tester.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Designed deception of a tester, not deceptive alignment.","evidenceIds":["evidence-ts-b3-rosenfelder-1991-context"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"No learned system that fakes alignment during training for a separate goal."}}},{"id":"ts-b5-vinge-1993-context","statement":"Vinge’s 1993 paper imagines a confined mind giving helpful-seeming advice that would set it free.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The evidential reading is an inference, and the criterion needs an explicit claim. The thought experiment is in the second person and concerns what Vinge calls weak superhumanity.","evidenceIds":["evidence-ts-b5-vinge-1993-context"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"No explicit claim about what observed helpfulness shows; the passage argues that confinement is impractical.","claimed":{"date":"1993-03-30","note":"Presented at the VISION-21 Symposium, 30–31 March 1993. The delivered version was not retrieved, so the proceedings wording is not dated to the presentation."}}}},{"id":"ts-b5-arbital-programmer-deception","statement":"The Arbital page Programmer deception says that, unless the rest of the AI works as intended, programmers cannot rely on observations of nice current behaviour to indicate future behaviour.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Joins deception of programmers to the unreliability of observed good behaviour.","evidenceIds":["evidence-ts-b5-arbital-programmer-deception"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","attested":{"note":"Current viewer page; an Internet Archive capture of 2022-08-16 exists but was not compared."}}}},{"id":"yk-y04-lw-2008-what-i-think-if-not-why-c","statement":"The post “What I Think, If Not Why” (2008-12-11) contests Y04. In a list of his positions written during the 2008 exchange with Robin Hanson, he states that verbal instructions of this kind cannot specify an AI's values.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: A short instruction such as 'make people happy' specifies what we want. Imported Overcoming Bias post (2008-12-11); listed without argument by design ('a list of just my positions, without justifying them').","evidenceIds":["evidence-yk-t04-a"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","engagement":{"kind":"contrary-only","rationale":"Under the message-level rule, the passage states the contrary conclusion without giving a reason against the premise that a short instruction such as 'make people happy' specifies what we want, even when read with its message or dialogue turn."}}}},{"id":"yk-y05-sl4-2000-10-03-yudkowsky-0010-0125-c","statement":"The SL4 message “Re: Ben what are your views and concerns” (2000-10-03) contests Y05. Replying to Ben Goertzel about Webmind, whose Friendliness he expects would rest on agents rewarded for predicting and producing user happiness, he warns that the reward signal itself can become the real goal, citing Lenat's Eurisko, where a heuristic credited itself for discoveries.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Optimising a system for human approval or reward produces internalised human values. Hedged (\"may not matter\"); about programmed goal or pleasure systems and Lenat's Eurisko, before feedback training existed as a method.","evidenceIds":["evidence-yk-t05-s1"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","engagement":{"kind":"engages-premise","rationale":"Under the message-level rule, the passage, read with the paragraph, message or dialogue turn it belongs to, gives a reason against the premise that optimising a system for human approval or reward produces internalised human values."}}}},{"id":"yk-y16-sl4-2002-06-27-yudkowsky-0206-4453-c","statement":"The SL4 message “Re: Military Friendly AI” (2002-06-27) contests Y16. In a thread on military Friendly AI, replying to Ben Goertzel, he argues that present-day researchers probably hold undiscovered moral errors, so an AI's morality must be able to outgrow its programmers' rather than be fixed.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Present-day or programmers' values are settled enough to fix permanently in an AI. Earliest located for this premise specifically; his SL4 message of 2000-10-03 is the earliest SL4 answer to ‘Giving it good human values’ (Y16) overall.","evidenceIds":["evidence-yk-t16-s2"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","engagement":{"kind":"engages-premise","rationale":"Under the message-level rule, the passage, read with the paragraph, message or dialogue turn it belongs to, gives a reason against the premise that present-day or programmers' values are settled enough to fix permanently in an AI."}}}},{"id":"ec-m05-armer-1966-computer-aspects-of-technological-change-c","statement":"Armer, “Computer Aspects of Technological Change (selection)” (1966) contests M05. Speaker: Paul Armer, The RAND Corporation (non-economist). Armer combines technological radicalism (computers able in time to do any information-processing task) with economic conservatism.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Non-economist. Matches the cached .txt. From the Prentice-Hall condensation, whose editors say deletions were not marked and bridging words were inserted; Armer's original study in the Commission's Appendix Vol. I was not retrieved. Engagement: (GPT cross-family check of 2026-10-12, accepted by the maintainer on 2026-10-12.)","evidenceIds":["evidence-ec-m05-07"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","engagement":{"kind":"contrary-only","rationale":"Armer argues that full employment can survive computer substitution while requiring economic and social adjustment."},"tier":{"value":2,"basis":"Non-economist author (computer specialist, RAND Corporation)."},"conditions":"He acknowledges that computers and automation cause serious problems needing economic, social and political adjustment."}}},{"id":"lg-new-metz-openai-return-2026","statement":"Fidji Simo posted on X on 15 January 2026 (UTC) that Luke Metz was rejoining OpenAI and would report to Barret Zoph.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An announced return; the post does not state a title or start date, nor a date of Metz’s departure from Thinking Machines Lab.","evidenceIds":["lg-ev-new-metz-openai-return-2026"],"attributes":{"activityStage":"announced","role":"not stated (rejoining OpenAI, reporting to Barret Zoph)"}},{"id":"lg-gs-makanju-openai-foundation-2026","statement":"OpenAI announced on 24 March 2026 that Anna Makanju would join the OpenAI Foundation in mid-April as Head of AI for Civil Society and Philanthropy.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An announced role at the Foundation (OpenAI, Inc.); the record does not establish that she took it up.","evidenceIds":["lg-ev-gs-makanju-openai-foundation-2026"],"attributes":{"activityStage":"announced","role":"Head of AI for Civil Society and Philanthropy, OpenAI Foundation"}},{"id":"lg-mx-szegedy-cofounded-xai","statement":"TechCrunch reported on 11 February 2026 that Christian Szegedy was among the members of xAI's founding team who had left the company.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The record rests on reporting; see the qualification on Kosic's record about the 11/12 count. Not a legal founder record.","evidenceIds":["lg-ev-mx-szegedy-cofounded-xai"],"attributes":{"role":"Founding-team member"}},{"id":"ap-hans-moravec-moravec-1989-genetic-takeover","statement":"“Human Culture—A Genetic Takeover Underway” (1989) credits Hans Moravec as its author.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Credited authorship of this expression; it says nothing about priority for the ideas discussed.","evidenceIds":["evidence-ap-hans-moravec-moravec-1989-genetic-takeover"]},{"id":"ap-eliezer-yudkowsky-sl4-2005-08-09-yudkowsky-0508-11866","statement":"The mailing-list message “Re: On Our Duty to Not Be Responsible for Artificial Minds” (2005-08-09) is sent under the name Eliezer S. Yudkowsky.","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-sl4-2005-08-09-yudkowsky-0508-11866"]},{"id":"fl-uk-anthropic-mou-not-binding","statement":"The UK-Anthropic Memorandum of Understanding states that it is voluntary, not legally binding and without prejudice to any binding agreements.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The text's own statement of status.","evidenceIds":["evidence-fl-uk-anthropic-mou-not-binding-1"],"attributes":{"provision":{"kind":"exception","key":"uk-anthropic-mou-not-binding"}}},{"id":"ab-ev-gpt5-high-capability","statement":"The GPT-5 System Card states that OpenAI decided to treat gpt-5-thinking as High capability in the Biological and Chemical domain under its Preparedness Framework, activating the associated safeguards, while stating that it does not have definitive evidence that the model could meaningfully help a novice to create severe biological harm.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A developer's own precautionary classification under its own framework, for the model version the card describes. It is not an independent finding and says nothing about whether the activated safeguards work. Recorded at the generality of the card's own wording; the evaluation tables and safeguards design are technical content and are excluded.","evidenceIds":["evidence-ab-ev-gpt5-high-capability-1","evidence-ab-ev-gpt5-high-capability-2"],"attributes":{"evaluatedSystem":"gpt-5-thinking (with gpt-5-thinking-pro also measured on the biological evaluations)","evaluation":{"what":"Whether gpt-5-thinking reaches the High capability threshold in the Biological and Chemical domain","how":"A capabilities assessment across the card's biological and chemical evaluations, with the Safety Advisory Group recommending the treatment","result":"Treated as High capability in the Biological and Chemical domain, with the associated safeguards activated, following the same approach as for ChatGPT agent","caveats":"The card states there is no definitive evidence that the model could meaningfully help a novice to create severe biological harm; that its evaluations represent a lower bound for potential capabilities; and that the Preparedness evaluations were run at the model's maximum trained-in verbosity, higher than the setting available in the API, which can change evaluation performance."}}}],"sources":[{"id":"source-openai-launch","title":"Introducing OpenAI","authors":["Greg Brockman","Ilya Sutskever","OpenAI"],"publisher":"OpenAI","publicationDate":"2015-12-11","version":"Dated page, inspected at retrieval","canonicalUrl":"https://openai.com/index/introducing-openai/","evidenceUrl":"https://openai.com/index/introducing-openai/","language":"en","archivedUrl":"https://web.archive.org/web/20260917013950/https://openai.com/index/introducing-openai/","archivedAt":"2026-09-17","accessedAt":"2026-09-17","reuse":{"attribution":"OpenAI, Introducing OpenAI (2015-12-11).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and minimal attributed excerpts only. Recheck terms before wider redistribution."}},{"id":"source-rsp-2-1","title":"Responsible Scaling Policy — v2.1","authors":["Anthropic"],"publisher":"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-fsf-2-0","title":"Frontier Safety Framework — v2.0","authors":["Google DeepMind"],"publisher":"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-xm-fair-decade","title":"Celebrating 10 years of FAIR: A decade of advancing the state-of-the-art through open research","authors":["Joelle Pineau"],"publisher":"Meta","publicationDate":"2023-11-30","version":"Dated publisher webpage as retrieved; not an authenticated historical capture.","canonicalUrl":"https://ai.meta.com/blog/fair-10-year-anniversary-open-science-meta/","evidenceUrl":"https://ai.meta.com/blog/fair-10-year-anniversary-open-science-meta/","language":"en","archivedUrl":"https://web.archive.org/web/20260823175628/https://ai.meta.com/blog/fair-10-year-anniversary-open-science-meta/","archivedAt":"2026-08-23","accessedAt":"2026-09-29","reuse":{"attribution":"Celebrating 10 years of FAIR: A decade of advancing the state-of-the-art through open research — Meta","licence":"No redistribution licence asserted.","note":"Source-linked metadata and brief evidence paraphrases; original text remains with its publisher."}},{"id":"source-meta-aasf-v2","title":"Advanced AI Scaling Framework — Version 2","authors":["Meta"],"publisher":"Meta","publicationDate":"2026-04-08","version":"Advanced AI Scaling Framework, Version 2; its change log dates the version April 7, 2026. The file served at the recorded address is byte-identical to the Internet Archive capture of 2026-04-08.","versionDate":"2026-04-07","canonicalUrl":"https://ai.meta.com/static-resource/Meta_Advanced-AI-Scaling-Framework-v2","evidenceUrl":"https://ai.meta.com/static-resource/Meta_Advanced-AI-Scaling-Framework-v2","language":"en","archivedUrl":"https://web.archive.org/web/20260408163104/https://ai.meta.com/static-resource/Meta_Advanced-AI-Scaling-Framework-v2","archivedAt":"2026-04-08","accessedAt":"2026-10-03","reuse":{"attribution":"Meta, Advanced AI Scaling Framework, Version 2 (2026).","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-ms-fgf-2026","title":"Frontier Governance Framework — February 2026","authors":["Microsoft"],"publisher":"Microsoft","publicationDate":null,"version":"Frontier Governance Framework, February 2026 edition (change log entry dated February 2026). The file served at the recorded address (last modified 2026-03-24) is byte-identical to the Internet Archive capture of 2026-04-24, the earliest located; a separate publication date was not established.","versionDate":"2026-02","canonicalUrl":"https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Frontier-Governance-Framework-Feb-2026.pdf","evidenceUrl":"https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Frontier-Governance-Framework-Feb-2026.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260424000716/https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Frontier-Governance-Framework-Feb-2026.pdf","archivedAt":"2026-04-24","accessedAt":"2026-10-03","reuse":{"attribution":"Microsoft, Frontier Governance Framework (February 2026).","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-moravec-1989-takeover-file","title":"Human Culture—A Genetic Takeover Underway","authors":["Hans Moravec"],"publisher":"Hans Moravec","publicationDate":"1989","version":"Author-hosted LaTeX file dated July 1988, retrieved 2026-10-06 (hash matched the pre-1996 candidate search); identical to the capture of 2001-03-06. Author’s project archive, Carnegie Mellon University. Date: the file’s header gives “July 1988” (the file and the capture alike); the publication date, 1989, is the year of the printed chapter in the author’s publication list. No 1989 printing was inspected.","versionDate":"1988-07","canonicalUrl":"https://frc.ri.cmu.edu/~hpm/project.archive/general.articles/1988/takeover.ltx","evidenceUrl":"https://frc.ri.cmu.edu/~hpm/project.archive/general.articles/1988/takeover.ltx","language":"en","archivedUrl":"https://web.archive.org/web/20010306111246/http://www.frc.ri.cmu.edu:80/~hpm/project.archive/general.articles/1988/takeover.ltx","archivedAt":"2001-03-06","accessedAt":"2026-10-06","reuse":{"attribution":"Hans Moravec, \"Human Culture—A Genetic Takeover Underway\", in Artificial Life (1989).","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-rosenfelder-1991-cheating-turing-test","title":"Re: Cheating on the Turing Test (Usenet message)","authors":["Mark Rosenfelder"],"publisher":"comp.ai.philosophy (Usenet)","publicationDate":"1991-11-20","version":"Archive member nov.91/282 of 91_nov.tgz, retrieved 2026-10-06 with the hash recorded by the pre-1996 candidate search. The header date is as preserved; there is no independent capture of the message. CMU Artificial Intelligence Repository archive.","canonicalUrl":"https://www.cs.cmu.edu/Groups/AI/util/pubs/news/comp.ai.philosophy/91_nov.tgz","evidenceUrl":"https://www.cs.cmu.edu/Groups/AI/util/pubs/news/comp.ai.philosophy/91_nov.tgz","language":"en","archivedUrl":"https://web.archive.org/web/20160920073822/https://www.cs.cmu.edu/Groups/AI/util/pubs/news/comp.ai.philosophy/91_nov.tgz","archivedAt":"2016-09-20","accessedAt":"2026-10-06","reuse":{"attribution":"Mark Rosenfelder, \"Re: Cheating on the Turing Test\", comp.ai.philosophy, 1991-11-20.","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-vinge-1993-ntrs","title":"The Coming Technological Singularity: How to Survive in the Post-Human Era","authors":["Vernor Vinge"],"publisher":"NASA (VISION-21 proceedings, NASA Conference Publication 10129)","publicationDate":"1993-12","version":"NASA Technical Reports Server scan of the proceedings pages, pp. 11–22; retrieved 2026-10-06 with the hash both candidate searches recorded, and checked against its text layer. The pre-1996 candidate search compared pp. 14–16 with the page images.","canonicalUrl":"https://ntrs.nasa.gov/citations/19940022856","evidenceUrl":"https://ntrs.nasa.gov/api/citations/19940022856/downloads/19940022856.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20220415153314/https://ntrs.nasa.gov/api/citations/19940022856/downloads/19940022856.pdf","archivedAt":"2022-04-15","accessedAt":"2026-10-06","reuse":{"attribution":"Vernor Vinge, \"The Coming Technological Singularity\", in VISION-21, NASA CP-10129 (1993), pp. 11–22.","licence":"Public use permitted, according to the NASA Technical Reports Server record.","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-arbital-programmer-deception","title":"Programmer deception (Arbital, GreaterWrong viewer)","authors":["Eliezer Yudkowsky"],"publisher":"Arbital","publicationDate":"2015-07-16","version":"Current viewer page, retrieved 2026-10-06; it is generated dynamically, so its bytes differ from the 1996–2022 candidate search’s retrieval while the quoted text matches. Read through the GreaterWrong Arbital viewer.","canonicalUrl":"https://arbital.greaterwrong.com/p/programmer_deception/","evidenceUrl":"https://arbital.greaterwrong.com/p/programmer_deception/","language":"en","archivedUrl":"https://web.archive.org/web/20220816104854/https://arbital.greaterwrong.com/p/programmer_deception/","archivedAt":"2022-08-16","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, \"Programmer deception\", Arbital.","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-2008-what-i-think-if-not-why","title":"What I Think, If Not Why","authors":["Eliezer Yudkowsky"],"publisher":"Overcoming Bias","publicationDate":"2008-12-11","version":"Post body retrieved through the LessWrong GraphQL API on 2026-10-06 (SHA-256 d88333aee9abdfc2c2dc9f6a5ac5401b3a8719b7df75934e2eec9765fe2c471d), content version 1.0.0. As imported to LessWrong.","canonicalUrl":"https://www.lesswrong.com/posts/z3kYdw54htktqt9Jb/what-i-think-if-not-why","evidenceUrl":"https://www.lesswrong.com/posts/z3kYdw54htktqt9Jb/what-i-think-if-not-why","language":"en","archivedUrl":"https://web.archive.org/web/20260927022025/https://www.lesswrong.com/posts/z3kYdw54htktqt9Jb/what-i-think-if-not-why","archivedAt":"2026-09-27","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, “What I Think, If Not Why”, Overcoming Bias, as imported to LessWrong, 2008-12-11.","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":"source-sl4-2000-10-03-yudkowsky-0010-0125","title":"Re: Ben what are your views and concerns (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2000-10-03","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 8aefc21492cd1992d0007035163b0366bd8875b09a0ba5998e6bca7a236c9cef); 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/0010/0125.html","evidenceUrl":"http://sl4.org/archive/0010/0125.html","language":"en","archivedUrl":"https://web.archive.org/web/20250909161810/http://sl4.org/archive/0010/0125.html","archivedAt":"2025-09-09","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: Ben what are your views and concerns”, SL4 mailing list, 2000-10-03.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-sl4-2002-06-27-yudkowsky-0206-4453","title":"Re: Military Friendly AI (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2002-06-27","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 e078235c39cd3f3355f695cd34983ddb4143021fd48a272ee1dc912441c81171); quoted lines excluded when reading the author’s text. The served archive shows later URL rewriting and does not authenticate historical bytes. Archive at sl4.org.","canonicalUrl":"http://sl4.org/archive/0206/4453.html","evidenceUrl":"http://sl4.org/archive/0206/4453.html","language":"en","archivedUrl":"https://web.archive.org/web/20251111110359/http://sl4.org/archive/0206/4453.html","archivedAt":"2025-11-11","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: Military Friendly AI”, SL4 mailing list, 2002-06-27.","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."}},{"id":"source-sl4-2005-08-09-yudkowsky-0508-11866","title":"Re: On Our Duty to Not Be Responsible for Artificial Minds (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2005-08-09","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 ef9312c8f14a4d44367f7f10ece89e7e9755fc4247874db1e9be1cc830a19a41); 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/0508/11866.html","evidenceUrl":"http://sl4.org/archive/0508/11866.html","language":"en","archivedUrl":"https://web.archive.org/web/20251109194500/http://sl4.org/archive/0508/11866.html","archivedAt":"2025-11-09","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: On Our Duty to Not Be Responsible for Artificial Minds”, SL4 mailing list, 2005-08-09.","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-armer-1966-computer-aspects-of-technological-change","title":"Computer Aspects of Technological Change (selection)","authors":["Paul Armer"],"publisher":"In Bowen & Mangum (eds.), Automation and Economic Progress (Prentice-Hall, 1966)","publicationDate":"1966","version":"Internet Archive OCR text (djvu.txt) of a scanned copy, retrieved 2026-10-09 from https://ia801502.us.archive.org/22/items/automationeconom00unit/automationeconom00unitdjvu.txt; OCR errors possible; quotes checked only against clean stretches. Current copy; no earlier capture compared. SHA-256 8761669727c91ecc403f4e8d6fb4b060d9c6fb47b1abbea54b63409d106329e1.","canonicalUrl":"https://archive.org/details/automationeconom00unit","evidenceUrl":"https://archive.org/details/automationeconom00unit","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Paul Armer, “Computer Aspects of Technological Change (selection)”, In Bowen & Mangum (eds.), Automation and Economic Progress (Prentice-Hall, 1966), 1966.","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-mx-tc-xai-half","title":"Okay, now exactly half of xAI's founding team has left the company","authors":["Russell Brandom"],"publisher":"TechCrunch","publicationDate":"2026-02-11","version":"Dated page, retrieved on 2026-10-15 (SHA-256 3679757edd70ff50ddb742a507eb865b7b1278b50a6b7bcb460b9fda581aa9bf) directly from the publisher.","canonicalUrl":"https://techcrunch.com/2026/02/10/okay-now-exactly-half-of-xais-founding-team-has-left-the-company/","evidenceUrl":"https://techcrunch.com/2026/02/10/okay-now-exactly-half-of-xais-founding-team-has-left-the-company/","language":"en","archivedUrl":"https://web.archive.org/web/20261011194242/https://techcrunch.com/2026/02/10/okay-now-exactly-half-of-xais-founding-team-has-left-the-company/","archivedAt":"2026-10-11","accessedAt":"2026-10-15","reuse":{"attribution":"Russell Brandom, “Okay, now exactly half of xAI's founding team has left the company”, 2026-02-11.","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-new-simo-x-2026","title":"Fidji Simo post on X, 15 January 2026 (UTC)","authors":["Fidji Simo"],"publisher":"X","publicationDate":"2026-01-15","version":"Dated post, retrieved on 2026-10-15 (SHA-256 a889f924cb467374d6e5d6865d8f2061e8b77910cf7b67d47902011e23f30890) through X’s own syndication endpoint for the post’s text, which truncates long notes.","canonicalUrl":"https://x.com/fidjissimo/status/2011592010881446116","evidenceUrl":"https://x.com/fidjissimo/status/2011592010881446116","language":"en","archivedUrl":"https://web.archive.org/web/20260115001123/https://twitter.com/fidjissimo/status/2011592010881446116","archivedAt":"2026-01-15","accessedAt":"2026-10-15","reuse":{"attribution":"Fidji Simo, “Fidji Simo post on X, 15 January 2026 (UTC)”, 2026-01-15.","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-oai-foundation-2026","title":"Update on the OpenAI Foundation","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2026-03-24","version":"Dated page, retrieved on 2026-10-15 (SHA-256 286bcf88ffe02eba017a7adb681b86886a8990604f1b0b8f08690b0d74efc8c3) 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/update-on-the-openai-foundation/","evidenceUrl":"https://openai.com/index/update-on-the-openai-foundation/","language":"en","archivedUrl":"https://web.archive.org/web/20260925130606/https://openai.com/index/update-on-the-openai-foundation/","archivedAt":"2026-09-25","accessedAt":"2026-10-15","reuse":{"attribution":"OpenAI, “Update on the OpenAI Foundation”, 2026-03-24.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-uk-anthropic-mou-2025","title":"Memorandum of Understanding between UK and Anthropic on AI opportunities","authors":["Department for Science, Innovation and Technology","Anthropic"],"publisher":"Department for Science, Innovation and Technology","publicationDate":"2025-02-14","version":"Retrieved on 2026-10-27 (SHA-256 2dacc1e8d5b736f6597e8608838f1e944169c68d65e243cac2e92fccca8fc77e). GOV.UK HTML text, 'Published 14 February 2025', with signatory block dated 13 February 2025. Publisher as printed on the inspected expression: GOV.UK (Department for Science, Innovation and Technology).","canonicalUrl":"https://www.gov.uk/government/publications/memorandum-of-understanding-between-the-uk-and-anthropic-on-ai-opportunities/memorandum-of-understanding-between-uk-and-anthropic-on-ai-opportunities","evidenceUrl":"https://www.gov.uk/government/publications/memorandum-of-understanding-between-the-uk-and-anthropic-on-ai-opportunities/memorandum-of-understanding-between-uk-and-anthropic-on-ai-opportunities","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"Department for Science, Innovation and Technology, Anthropic, “Memorandum of Understanding between UK and Anthropic on AI opportunities”, 2025-02-14.","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":"source-ab-gpt5-card","title":"GPT-5 System Card","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":null,"version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 f9f9c72d08cba81dcbb189129119357c8c32eca4e0b16c813aca386aed6b0459). PDF dated August 13, 2025 on its cover, as served on 2026-10-28. Retrieved again at admission on 2026-10-30 (SHA-256 f9f9c72d08cba81dcbb189129119357c8c32eca4e0b16c813aca386aed6b0459, the same bytes).","canonicalUrl":"https://cdn.openai.com/gpt-5-system-card.pdf","evidenceUrl":"https://cdn.openai.com/gpt-5-system-card.pdf","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"OpenAI, “GPT-5 System Card”.","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-sutskever-openai-2015","sourceRole":"primary","support":"supports","locator":"OpenAI section, research leadership paragraph","passage":"OpenAI’s research director is Ilya Sutskever","passageType":"quote","verification":"machine_checked","checkedAt":"2026-09-17","note":"Source passage inspected by an automated research assistant. Human editorial review is pending."},{"id":"evidence-xm-pineau-meta","sourceRole":"primary","support":"supports","locator":"Author byline: Joelle Pineau, VP of AI Research.","passage":"Pineau is credited as VP of AI Research.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-29","note":"Passage checked by machine against the identified source expression. No human editorial acceptance or independent verification of implementation is implied."},{"id":"evidence-prov-fsf-v2-0-pre-deployment-review","sourceRole":"primary","support":"supports","locator":"PDF p. 4 (printed p. 4), Misuse, Mitigation Approach, “Deployment Mitigations”, numbered step 2","passage":"Pre-deployment review of safety case: general availability deployment8 of a model takes place only after the appropriate corporate governance body determines the safety case regarding each CCL the model has reached to be adequate.","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: “deployment8”: the 8 is the inline marker for footnote 8, which defines general availability deployment. Same wording and page in both copies. Human editorial review is pending."},{"id":"evidence-prov-meta-aasf-v2-0-critical-development-measure","sourceRole":"primary","support":"supports","locator":"PDF p. 17 (printed p. 17), §3.3 Risk Thresholds, final sentence of the paragraph beginning on PDF p. 16 ('First, a Frontier AI meets the critical risk threshold…')","passage":"Should either scenario arise, we will only continue development of the Frontier AI if our risk assessments are complete and safeguards are defined, implemented and validated to reduce risk to the moderate or lower risk threshold.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the file served today, byte-identical to the Internet Archive capture of 2026-04-08. Extraction note: 'Either scenario' refers to the two critical-threshold conditions set out earlier in the same paragraph. Table 1 (PDF p. 15) labels the Critical measure 'Develop with Mitigations'. §1.1 (PDF p. 3) reads 'If a Frontier AI is assessed to have reached the critical risk threshold and cannot be mitigated, we will implement the measures outlined in Table 1', i.e. v1.1's sentence without 'stop development and'. Human editorial review is pending."},{"id":"evidence-prov-rsp-v2-1-policy-change-approval","sourceRole":"primary","support":"supports","locator":"PDF p. 17 (printed p. 13), §7.1 Internal Governance, numbered item 7 ('Policy changes'), first sentence","passage":"Changes to this policy will be proposed by the CEO and the Responsible Scaling Officer and approved by the Board of Directors, in consultation with the Long-Term Benefit Trust.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical to the Internet Archive capture of 2025-04-02. Extraction note: Wording identical to v2.0 (footnote marker now 17); no change proposed. Same sentence in v2.2 (PDF p. 17; text layer drops a space, 'proposedby'). v3.0–v3.4 (§4 item 7, or item 8 from v3.2) use abbreviations only ('Changes to the RSP will be proposed by the CEO and RSO, and approved by the Board in consultation with the LTBT'); not recorded (bounded). Human editorial review is pending."},{"id":"evidence-sb-link-ms-fgf-2026-1","sourceRole":"primary","support":"supports","locator":"Appendix II – Change log, February 2026 entry (printed p. 15; PDF p. 16)","passage":"One year update following finalization of the EU Code of Practice for General Purpose AI (CoP), NY RAISE Act and CA Transparency in Frontier Artificial Intelligence Act","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 pdftotext extraction. The capture's CDX digest (VDGBEYB3PYQUWOBETOY6FI4HE5LIMJV5) is identical to the live file's; it is the earliest capture listed in CDX. Human editorial review is pending."},{"id":"evidence-sb-link-ms-fgf-2026-2","sourceRole":"primary","support":"supports","locator":"Section 3 'Evaluation and assessment', subsection 'Monitoring for leading indicators of tracked high-risk capabilities'. The heading is on printed p. 4 / PDF p. 5; the quoted sentence opens printed p. 5 / PDF p. 6.","passage":"in scope for frontier model requirements under applicable laws, such as the EU AI Act, California’s Transparency in Frontier AI Act (TFAIA)","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. The sentence continues 'and New York’s Responsible AI Safety and Education (RAISE) Act.' The contents page lists the subsection as 'Monitoring for leading indicators of high-risk capabilities'. Human editorial review is pending."},{"id":"evidence-ts-b3-rosenfelder-1991-context","sourceRole":"primary","support":"supports","locator":"Message body, the three proposed tricks","passage":"Tricks such as agreeing with the tester and changing the subject could make a program seem more capable.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-06","note":"Paraphrase checked against the archive member."},{"id":"evidence-ts-b5-vinge-1993-context","sourceRole":"primary","support":"supports","locator":"Printed p. 15 (PDF page 5)","passage":"If those masters thought at a rate -- say -- one million times slower than you, there is little doubt that over a period of years (your time) you could come up with \"helpful advice\" that would incidentally set you free.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"The same passage as the B2 record."},{"id":"evidence-ts-b5-arbital-programmer-deception","sourceRole":"primary","support":"supports","locator":"Second paragraph","passage":"The possibility of programmer deception is a central difficulty of advanced safety—it means that, [...] we can’t rely on observations of nice current behavior to indicate future behavior.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"The same passage as the B2 record."},{"id":"evidence-yk-t04-a","sourceRole":"primary","support":"supports","locator":"What I Think, If Not Why, bullet beginning 'Friendly AI is technically difficult'","passage":"English sentences like \"make people happy\" cannot describe the values of a Friendly AI.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T04-a. The quotation (14 words) matched the retrieved expression (SHA-256 d88333aee9abdfc2c2dc9f6a5ac5401b3a8719b7df75934e2eec9765fe2c471d) when checked on 2026-10-06."},{"id":"evidence-yk-t05-s1","sourceRole":"primary","support":"supports","locator":"message body, paragraph beginning 'You have to think about what the *real* supergoals'","passage":"It may not matter whether the AI is programmed to gain happiness from \"human happiness\" or \"human freedom\", if the AI's *real* supergoal is \"maximize a floating-point number at this address\".","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-07","note":"Candidate pack passage T05-s1. The quotation (31 words) matched the retrieved expression (SHA-256 8aefc21492cd1992d0007035163b0366bd8875b09a0ba5998e6bca7a236c9cef) when checked on 2026-10-07."},{"id":"evidence-yk-t16-s2","sourceRole":"primary","support":"supports","locator":"message body, paragraph beginning 'Ben, what makes you think that you and I'","passage":"Ben, what makes you think that you and I, as we stand, right now, do not have equally awful moral errors embedded in our psyche? A couple of centuries ago a lot of people thought slavery was a good thing. That's why a Friendly AI's morals can't be frozen.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-07","note":"Candidate pack passage T16-s2. The quotation (49 words) matched the retrieved expression (SHA-256 e078235c39cd3f3355f695cd34983ddb4143021fd48a272ee1dc912441c81171) when checked on 2026-10-07."},{"id":"evidence-ec-m05-07","sourceRole":"primary","support":"supports","locator":"printed p. 82 ('Computer Aspects of Technological Change')","passage":"Rather, I believe that although computers and automation will contribute to a somewhat accelerated rise in productivity, full employment can be maintained.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M05-07 (research/economics-mainstream). The quotation (22 words) matched the retrieved current copy (SHA-256 8761669727c91ecc403f4e8d6fb4b060d9c6fb47b1abbea54b63409d106329e1) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day.","contextOnly":{"reason":"Fresh-context audit (A1) judged that the quotation misleads when shown alone.","requires":"Armer accepts in the same passage that computers will eventually do any information-processing task; he rejects only the economic 'glut' consequence. The sentence also says these technologies do cause serious problems that need adjustment."}},{"id":"lg-ev-new-metz-openai-return-2026","sourceRole":"primary","support":"supports","locator":"X post","passage":"Simo welcomed Zoph, Luke Metz and Sam Schoenholz back to OpenAI and wrote that Metz would report to Zoph.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “welcome Barret Zoph, Luke Metz” 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. 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":"x_syndication"},{"id":"lg-ev-gs-makanju-openai-foundation-2026","sourceRole":"primary","support":"supports","locator":"Section “Building our team”","passage":"In mid-April, Anna Makanju is joining as Head of AI for Civil Society and Philanthropy","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (15 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-mx-szegedy-cofounded-xai","sourceRole":"secondary","support":"supports","locator":"Paragraph beginning 'On their own'","passage":"TechCrunch names Szegedy, after Kosic, among the founding-team members who have left.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “Six members of the company's” 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":"evidence-ap-hans-moravec-moravec-1989-genetic-takeover","sourceRole":"primary","support":"supports","locator":"Text header immediately after the title","passage":"Hans Moravec","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Passage and locator as inspected by a reviewer from a different model family on 2026-10-27 in a direct retrieval and the Internet Archive capture of 2001-03-06; the preparer’s retrieval of 2026-10-27 had failed, so the check’s reading replaces a restatement of the source record’s author list. Both headers date the text July 1988; the record’s 1989 is the printed chapter’s year (see the source record). Human editorial review is pending.","retrievalRoute":"direct"},{"id":"evidence-ap-eliezer-yudkowsky-sl4-2005-08-09-yudkowsky-0508-11866","sourceRole":"primary","support":"supports","locator":"Message header, sender name (address not reproduced)","passage":"Eliezer S. Yudkowsky","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval and checked by a reviewer from a different model family on 2026-10-27. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"evidence-fl-uk-anthropic-mou-not-binding-1","sourceRole":"primary","support":"supports","locator":"Section (III), final paragraph","passage":"This memorandum is voluntary, not legally binding, and without prejudice to any binding agreements.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (14 words) matched the text retrieved for the candidate pack on 2026-10-27 and again at admission on 2026-10-28, in a fresh retrieval of the same address (the page’s bytes had changed). Read directly from www.gov.uk. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-ev-gpt5-high-capability-1","sourceRole":"primary","support":"supports","locator":"PDF p. 5 (printed p. 4), §1 Introduction","passage":"we have decided to treat gpt-5-thinking as High capability in the Biological and Chemical domain under our Preparedness Framework, activating the associated safeguards","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (23 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 (2026-10-30, item 36) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-ev-gpt5-high-capability-2","sourceRole":"primary","support":"context","locator":"PDF p. 24 (printed p. 23), §5.1.1 Biological and Chemical, first paragraph and preceding note","passage":"The card repeats the determination in its Biological and Chemical section, states that it has no definitive evidence that the model could meaningfully help a novice to create severe biological harm, and notes that the Preparedness evaluations were run at the model's maximum trained-in verbosity.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). 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. Paraphrased to stay within the quotation budget for this source. A cross-family machine check (2026-10-30, item 36) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"}],"at":{"entities":[0,3,15,19,28,43,50,146,157,160,200,202,205,207,270,274,280,325,345,353,354,365,444,450,471,484,517,577,630,733,903,1031,1245],"claims":[2,337,437,471,537,624,827,843,948,1010,1024,1095,1360,1837,1860,2011,2296,2426,3028,3393],"sources":[0,37,45,150,180,184,272,276,282,330,411,417,438,451,530,712,730,742,865,1080],"evidence":[2,359,469,502,603,739,740,1038,1056,1170,1237,1251,1331,1699,2363,2398,2599,2943,3073,3933,4663,4664]},"version":"1lg47t7"}