{"entities":[{"id":"demis-hassabis","name":"Demis Hassabis"},{"id":"good-1965","name":"Speculations Concerning the First Ultraintelligent Machine"},{"id":"rsp-v3-3","name":"Anthropic Responsible Scaling Policy — v3.3"},{"id":"rsp-v3-4","name":"Anthropic Responsible Scaling Policy — v3.4"},{"id":"fsf-v3-0","name":"Google DeepMind Frontier Safety Framework — v3.0"},{"id":"fsf-v3-1","name":"Google DeepMind Frontier Safety Framework — v3.1"},{"id":"preparedness-beta-2023","name":"OpenAI Preparedness Framework — Beta"},{"id":"frontier-governance-2026","name":"OpenAI Frontier Governance Framework — May 2026 edition"},{"id":"faif-2026","name":"xAI Frontier Artificial Intelligence Framework — June 2026 edition"},{"id":"zhipu","name":"Zhipu AI / 智谱"},{"id":"wang-shaolan","name":"Wang Shaolan / 王绍兰"},{"id":"meta-aasf-v2-0","name":"Meta Advanced AI Scaling Framework — Version 2"},{"id":"us-white-house","name":"The White House"},{"id":"butler-1865-lucubratio-ebria","name":"Lucubratio Ebria"},{"id":"vinge-1993-coming-technological-singularity","name":"The Coming Technological Singularity: How to Survive in the Post-Human Era"},{"id":"hanson-2008-what-core-argument","name":"What Core Argument?"},{"id":"agi-ruin-2022","name":"AGI Ruin: A List of Lethalities"},{"id":"seed-ai","name":"Seed AI"},{"id":"fast-or-hard-takeoff","name":"Fast or hard takeoff"},{"id":"ai-automated-ai-research","name":"AI-automated AI research"},{"id":"control-or-purpose-specification","name":"Control or purpose specification"},{"id":"diminishing-returns-to-intelligence","name":"Diminishing returns rule out a runaway"},{"id":"human-data-humanlike-mind","name":"Trained on human data, so humanlike"},{"id":"no-body-no-danger","name":"Without a body it cannot do much"},{"id":"extropians-1998-09-18-yudkowsky-2487","name":"Re: Singularity: The Moore's Law Argument"},{"id":"lw-2023-gpts-are-predictors-not-imitators","name":"GPTs are Predictors, not Imitators"},{"id":"ai-general-purpose-technology-with-lag","name":"AI as a general-purpose technology with lagging productivity"},{"id":"compensating-policy-for-ai","name":"Compensating policy for AI’s distributional effects"},{"id":"sachs-kotlikoff-2012-smart-machines","name":"Smart Machines and Long-Term Misery"},{"id":"furman-2019-automation-future-past","name":"Should We Be Reassured If Automation in the Future Looks Like Automation in the Past?"},{"id":"ai-existential-risk-in-economic-analysis","name":"Existential risk from AI as a matter for economic analysis"},{"id":"jones-2024-ai-dilemma-aer-insights","name":"The AI Dilemma: Growth versus Existential Risk"},{"id":"alphabet","name":"Alphabet"},{"id":"us-eo-14409-2026","name":"Executive Order 14409: Promoting Advanced Artificial Intelligence Innovation and Security"},{"id":"us-eo-14292-2025","name":"Executive Order 14292: Improving the Safety and Security of Biological Research"}],"claims":[{"id":"fsf-v3-1-cbrn-tcls","statement":"FSF v3.1 introduces CBRN Tracked Capability Levels.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"This is a scoped change in the policy text. It does not establish implementation, compliance, risk reduction or a comparative safety ranking.","evidenceIds":["evidence-fsf-v3-1-cbrn-tcls"],"attributes":{"changeBasis":"changelog"}},{"id":"absent-faif-2026-benefits-may-outweigh-benchmark-risks","statement":"No passage stating the provision “Deployment benefits may outweigh benchmark-identified risks” was found in FAIF · June 2026 edition within the inspected scope.","derivation":"inferred","resolution":"supported","inferenceRationale":"Premises: FAIF · December 2025 edition states this provision. The inspected scope of FAIF · June 2026 edition was: June 2026 edition, PDF pp. 1–9, whole document: §1 Introduction; §2 Risk Management Framework with §2.1 (a)–(c), §2.2, both sections numbered 2.3 and §2.4; §3 Incident reporting; §4 Governance. Method: Read the whole document, in particular both sections numbered 2.3 (Systemic risk acceptance determination; Safety mitigations); searched the text layer for “outweigh”, “conceivable”, “benefit” and “balance”. No passage stating that the expected benefits of deployment may outweigh the risks identified by a particular benchmark was found within PDF pp. 1–9. 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. Text comparison only; no replacement between the editions is asserted by this record.","evidenceIds":["evidence-absent-faif-2026-benefits-may-outweigh-benchmark-risks"],"attributes":{"provision":{"kind":"exception","key":"faif-benefits-may-outweigh-benchmark-risks"}}},{"id":"prov-rsp-v3-4-ltbt-review-request","statement":"RSP v3.4 states that, upon the LTBT's request, Anthropic will conduct a public or private external review of its Risk Report or sections of it, subject to the procedures outlined in the policy.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the edition states, quoted from the cited expression. It is not evidence of implementation, compliance or effectiveness.","evidenceIds":["evidence-prov-rsp-v3-4-ltbt-review-request"],"attributes":{"provision":{"kind":"decision_authority","key":"rsp-ltbt-review-request","parties":[{"role":"decides","actor":"Long-Term Benefit Trust"}]}}},{"id":"provchange-rsp-v3-4-automated-rd-threshold","statement":"v3.4 counts the dramatic-acceleration scenario of the Automated R&D threshold as having occurred where the observed or expected doubling of AI progress is measured against both the rate Anthropic would expect and the fastest rate of extended progress it has observed (over at least three model generations) without significant AI contributions to AI R&D, and where continuing the trend “seems likely to lead to” even greater acceleration; v3.3 measured the doubling against the expected rate alone, and continuing the trend “could lead to” even greater acceleration.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A textual difference between the two editions’ wording of this keyed provision. It is not an assessment of whether the change strengthened or weakened the framework, and not evidence of implementation.","evidenceIds":["evidence-provchange-rsp-v3-4-automated-rd-threshold-later","evidence-provchange-rsp-v3-4-automated-rd-threshold-earlier","evidence-provchange-rsp-v3-4-automated-rd-threshold-changelog"],"attributes":{"changeBasis":"changelog","provision":{"kind":"threshold","key":"rsp-automated-rd-threshold"}}},{"id":"ts-b6-butler-1865-contests","statement":"The unsigned 1865 article \"Lucubratio Ebria\" rejects a previous correspondent’s view that machines are separate beings whose final triumph over mankind should be expected, and treats them as the means by which the human organism develops.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The article frames itself as a dream after \"a single glass of hot whisky and water\" and closes by leaving \"our dream\", which bears on reading the reply as Butler’s own position. Its attribution to Butler rests on its close match with the limbs passage of Erewhon (1872); Butler’s 1901 preface mentions an untitled second Press article that appeared \"shortly after the first\", whereas this one appeared two years later. It attributes the growing change in people’s physical status to people with mechanical limbs, not to machines.","evidenceIds":["evidence-ts-b6-butler-1865-contests","evidence-ts-b6-butler-1865-contests-2"],"attributes":{"thought":{"roles":["contesting"],"register":"dream","qualifier":"Disputes the premise that machines are a separate kind of being that could triumph over people.","engagement":{"kind":"engages-premise","rationale":"Gives a reason against the premise that machines are a separate kind of being that could escape and triumph over people: they are the means by which the human organism itself develops."}}}},{"id":"ts-a5-good-1965","statement":"Good’s chapter counts the design of machines among the intellectual activities an ultraintelligent machine would surpass, so that machines would design better machines.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A design-feedback argument, not a description of any implemented system.","evidenceIds":["evidence-ts-a5-good-1965"],"attributes":{"thought":{"roles":["earliest-located"],"register":"nonfiction","claimed":{"date":"1962-10","note":"The chapter’s footnote on p. 31 names talks in October 1962 and January 1963, a first draft of April 1963 and a slightly emended version of May 1964. None was retrieved, so the wording is not dated to them."}}}},{"id":"ts-a4-vinge-1993","statement":"Vinge’s 1993 paper says the intellectual runaway will probably occur faster than any earlier technical revolution, and that a month or two, or a day or two, after it we will be in the post-human era.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A speculative timescale for the period after the event, not a measured forecast. The paper does not use the terms \"hard takeoff\" or \"fast takeoff\".","evidenceIds":["evidence-ts-a4-vinge-1993","evidence-ts-a4-vinge-1993-2"],"attributes":{"thought":{"roles":["earliest-located"],"register":"nonfiction","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-a2-hanson-2008-core-argument-contests","statement":"Hanson’s post summarises Yudkowsky’s scenario of a seed AI growing within roughly a week from unimportant to strong enough to take over the world, and rejects that growth rate as far beyond what most AI researchers, growth economists or other specialists would estimate.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Hanson’s own argument against the speed of the scenario. He does not dispute that an AI could be designed to improve its own design; his reasons bear on the pace of takeoff (A4), not on the seed-AI criterion.","evidenceIds":["evidence-ts-a2-hanson-2008-core-argument-contests"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","qualifier":"Rejects the growth rate of the week-long seed-AI scenario.","missing":"Names a seed AI architecture only to dispute how fast it could grow; it neither asserts nor disputes an initial AI designed to improve its own design.","attested":{"note":"Wording checked in the current re-hosted page and in MIRI’s 2013 collection; no 2008 capture was compared."}}}},{"id":"yk-y10-extropians-1998-09-18-yudkowsky-2487-c","statement":"The Extropians message “Re: Singularity: The Moore's Law Argument” (1998-09-18) contests Y10. Replying to Robin Hanson's critique of the 'Moore's Law argument', he concedes that the 'Recursive Moore's Law' section of 'Staring Into the Singularity' was an illustration, not a technical argument, but maintains that technologically enhanced intelligence creates positive feedback.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Returns to reinvesting cognitive gains in further cognitive improvement diminish. The opening paragraph's concession bears on the passage in 'Staring into the Singularity': he calls it an introduction or illustration of explosive growth 'not a technical argument in favor of it'. The same message disowns the 1996 recursive-Moore's-Law scenario as a technical argument.","evidenceIds":["evidence-yk-t10-e3"],"attributes":{"thought":{"roles":["contesting","later-restatement"],"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 returns to reinvesting cognitive gains in further cognitive improvement diminish."}}}},{"id":"yk-y21-lw-2023-gpts-are-predictors-not-imitators-c","statement":"The post “GPTs are Predictors, not Imitators” (2023-04-08) contests Y21. He argues that predicting human text requires modelling the processes behind it, a different and harder task than being human, so a predictor need not think like the humans it predicts.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Training on human text or behaviour yields human-like thought and motives. The post says related text was posted to Twitter and this version is edited. Elided words: 'even leaving aside all the ways that gradient descent differs from natural selection'.","evidenceIds":["evidence-yk-t21-a"],"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 training on human text or behaviour yields human-like thought and motives."}}}},{"id":"yk-y26-agi-ruin-2022-c","statement":"The post “AGI Ruin: A List of Lethalities” (2022-06-05) contests Y26. Opens an item that gives a 'lower-bound model' in which an AI with Internet access orders proteins by email and bribes or persuades an unwitting human to mix them, leading to nanomachinery.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Intelligence without physical actuators of its own is not dangerous.","evidenceIds":["evidence-yk-t26-c"],"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 intelligence without physical actuators of its own is not dangerous."}}}},{"id":"ec-m03-furman-2019-automation-future-past-b","statement":"Furman, “Should We Be Reassured If Automation in the Future Looks Like Automation in the Past?” (2019-05) bears on M03. Furman attributes AI's small macroeconomic impact to the newness of recent advances and notes that inventions like electricity added productivity in successive waves.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Describes a lag between recent advances and macro impact.","evidenceIds":["evidence-ec-m03-09"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Does not call AI a general-purpose technology or attribute the lag to complementary investment.","tier":{"value":1,"basis":"Chapter of the NBER Economics of AI volume (core venue)."}}}},{"id":"ec-m06b-sachs-kotlikoff-2012-smart-machines-a","statement":"Sachs and Kotlikoff, “Smart Machines and Long-Term Misery” (2012-12) articulates M06b. Rejecting 'smashing the machines', the authors propose intergenerational tax-and-transfer policy (in the model, a wealth levy on the old funding transfers to the young) to make all generations gain.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"NBER Working Paper 18629, dated December 2012, revised January 2013; a highly stylised two-period OLG model, not an empirical estimate or forecast. The policy is shown to work in a numerical illustration of the model, not evaluated empirically.","evidenceIds":["evidence-ec-m06b-09"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","attested":{"date":"2013-01","note":"The cover lists \"December 2012, January 2013\"; the retrieved PDF may be the January 2013 revision, so its wording is attested from 2013-01 only."},"tier":{"value":1,"basis":"NBER working paper (core venue); both authors economists."},"conditions":"Model illustration: unanticipated permanent rise in u from 1 to 10; government levies part of the machine stock and transfers its returns."}}},{"id":"ec-e03-jones-2024-ai-dilemma-aer-insights-a","statement":"Jones, “The AI Dilemma: Growth versus Existential Risk” (2024-12) articulates E03. Published version of Jones's working paper; the abstract attributes existential risk from a misaligned superintelligence to 'many experts'.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"AER: Insights, December 2024; abstract page only. The working paper (November 2023) carries the same sentence with 'A.I.' spelling and is the earlier dated text.","evidenceIds":["evidence-ec-e03-16"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","tier":{"value":1,"basis":"AER: Insights (AEA journal); Jones is an economist."}}}},{"id":"lg-cn-wang-shaolan-zhipu-deputy-gm-2025","statement":"Zhipu’s Hong Kong prospectus of 30 December 2025 identifies Wang Shaolan as deputy general manager.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A role shown on a dated filing. Read with the separate record of Wang Shaolan as CEO in June 2021, it shows a different title in December 2025; the filing does not date or explain the change, and none is inferred.","evidenceIds":["lg-ev-cn-wang-shaolan-zhipu-deputy-gm-2025"],"attributes":{"role":"deputy general manager"}},{"id":"lg-goo-hassabis-alphabet-chief-scientist-2026","statement":"Sundar Pichai announced on 5 August 2026 that Demis Hassabis would become Chief Scientist of Alphabet.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An announced role; the record does not establish a start date. The undated Keyword author page retrieved on 15 October 2026 shows the title (context only).","evidenceIds":["lg-ev-goo-hassabis-alphabet-2026","lg-ev-goo-hassabis-alphabet-author-page"],"attributes":{"activityStage":"announced","role":"Chief Scientist of Alphabet"}},{"id":"cmp-fsf-v3-1-d3-halt","statement":"FSF · v3.1 states a provision on the comparison question “Pause, non-deployment or non-training”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A deployment gate: external deployment (and, under §3.1.2, high-risk internal deployment) takes place only after a governance determination that residual risk is acceptable. The misuse deployment-mitigation acceptance criterion is 'required only for external deployment, not internal deployment or further development' (PDF p. 7). No passage states that development will be put on hold. 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-v3-1-d3-halt-1"],"attributes":{"provision":{"kind":"required_action","key":"comparison-d3-halt","facet":{"id":"D3.halt","status":"stated"}}}},{"id":"cmp-faif-2026-d6-discretion","statement":"FAIF · June 2026 edition states a provision on the comparison question “Other exceptions or discretion”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The December 2025 trusted-user and benefits-outweigh-benchmark clauses do not appear in this edition. 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-faif-2026-d6-discretion-1","ev-cmp-faif-2026-d6-discretion-2"],"attributes":{"provision":{"kind":"exception","key":"comparison-d6-discretion","facet":{"id":"D6.discretion","status":"stated"}}}},{"id":"cmp-meta-aasf-v2-0-d6-competitor","statement":"No passage meeting the comparison question “Provisions conditioned on other developers” was found in Meta AASF · v2 within the inspected scope.","derivation":"inferred","resolution":"supported","inferenceRationale":"Premises: the comparison question asks for the following. A provision that relaxes, adjusts or conditions the developer’s safeguards or commitments on what other developers do or have (often called a “race” or “marginal risk” clause). Scope tests that decide whether a model is covered by the framework are recorded as context, not counted. The inspected scope of Meta AASF · v2 was: Whole text layer, PDF pp. 1–44 (p. 45 blank): all sections, tables, footnotes and both appendices read. Method: the edition was read in full and its text layer searched case-insensitively for other developer / other compan, competit, race, marginal, industry, most advanced models, open source / widely available / publicly available. Hits examined and not counted: No hits for 'other developer', 'other compan', 'race' or 'marginal'. 'competition' (PDF pp. 28–29) names capture-the-flag competitions used as cyber benchmarks. 'industry' (PDF pp. 7, 8, 41) refers to evolving industry practices as a reason to re-evaluate mitigations, update the Framework or revise definitions. 'the most advanced models' (Appendix I, PDF p. 41; change log, PDF p. 44) is part of the scope criterion for which models count as Frontier AI: a model 'reasonably likely to be more capable … than our existing models or the most advanced models'. It decides which models the Framework covers, not what Meta does depending on other developers' conduct. PDF p. 4, §2.1.1: 'We compare these various factors against our own models and those available externally' identifies a 'reference class' of comparable models for tracking performance and anticipating risks and required assessments; it does not relax, adjust or condition Meta's safeguards or commitments. 'open-source' and 'widely used' (PDF pp. 5, 8, 31) concern evaluations and defender access. No provision relaxes, adjusts or conditions Meta's safeguards or commitments on what other developers do or have. The conclusion is bounded by that scope and method.","editorialState":"seed","qualification":"Comparative scope and reference-class provisions, not a waiver of safeguards conditional on other developers. Appendix I compares a model with 'the most advanced models' (not limited to Meta's) to decide whether it is Frontier AI (PDF p. 41); §2.1.1 compares a planned model 'against our own models and those available externally' to identify a 'reference class' used to anticipate risks and required assessments (PDF p. 4). Both are recorded here as context. A recorded absence within the stated scope: not a claim that the provision was deleted, and not a statement about practice or intent.","evidenceIds":["ev-cmp-meta-aasf-v2-0-d6-competitor"],"attributes":{"provision":{"kind":"exception","key":"comparison-d6-competitor","facet":{"id":"D6.competitor"}}}},{"id":"cmp-preparedness-beta-2023-d1-cyber","statement":"Preparedness · Beta states a provision on the comparison question “Cyber operations”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Named as a Tracked Risk Category with a Low to Critical table (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-preparedness-beta-2023-d1-cyber-1"],"attributes":{"provision":{"kind":"risk_domain","key":"comparison-d1-cyber","facet":{"id":"D1.cyber","status":"stated"}}}},{"id":"cmp-frontier-governance-2026-d7-internal","statement":"FGF · May 2026 edition states a provision on the comparison question “Internal reporting”, in a weaker form than the question’s criterion.","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. The passage records a documented justification but does not name its internal recipients. §2.6 states that the incident response plan includes notifying appropriate internal teams (PDF p. 14). 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-frontier-governance-2026-d7-internal-1"],"attributes":{"provision":{"kind":"reporting","key":"comparison-d7-internal","facet":{"id":"D7.internal","status":"partly"}}}},{"id":"ab-scr-eo14292-signed","statement":"Executive Order 14292, Improving the Safety and Security of Biological Research, is dated 5 May 2025 in the Federal Register print of the order.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Records the order's date as printed. Signature is not implementation: nothing here establishes that any step the order directs was taken. The order's gain-of-function research provisions are outside this pass.","evidenceIds":["evidence-ab-scr-eo14292-signed-1"]},{"id":"ab-gov-us-eo14409-no-biological-provision","statement":"Within the inspected scope, a case-insensitive search of the Federal Register text of Executive Order 14409 of 2 June 2026, as cached on 28 October 2026, finds no occurrence of 'biological', 'biosecurity', 'CBRN', 'chemical', 'pathogen', 'nucleic' or 'weapon'.","derivation":"inferred","resolution":"supported","inferenceRationale":"Premises: the inspected scope was the Federal Register text of Executive Order 14409 (91 FR 34565-34567), as served by govinfo.gov on 2026-10-28 (whole text, 1,317 words of text layer). Method: case-insensitive search for each term named in the statement, each returning zero matches, and reading of the order's purpose and of Sections 2 and 3, whose provisions concern cyber defence of systems, an AI cybersecurity clearinghouse and a classified benchmarking process for advanced cyber capabilities. The record means not found within that scope; it says nothing about other orders, about measures taken under this one, or about any classified annex.","editorialState":"seed","qualification":"A bounded string search of one expression. It is not a claim that the administration took no action on AI and biological risk, and it does not characterise the order's purpose beyond what the text states.","evidenceIds":["evidence-ab-gov-us-eo14409-no-biological-provision-1"],"attributes":{"provision":{"kind":"risk_domain","key":"us-eo14409-biological-risk-domain"}}}],"sources":[{"id":"source-good","title":"Speculations Concerning the First Ultraintelligent Machine","authors":["I. J. Good"],"publisher":"Academic Press","publicationDate":"1965","version":"Scanned reprint, Advances in Computers volume 6, pp. 31–88. Repository metadata identifies ©1965; the mirror filename says 1966.","canonicalUrl":"https://gwern.net/doc/ai/1966-good.pdf","evidenceUrl":"https://gwern.net/doc/ai/1966-good.pdf","language":"en","accessedAt":"2026-09-17","reuse":{"attribution":"Academic Press, Speculations Concerning the First Ultraintelligent Machine (1965).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and minimal attributed excerpts only. Recheck terms before wider redistribution."}},{"id":"source-rsp-3-3","title":"Responsible Scaling Policy — v3.3","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":null,"version":"Responsible Scaling Policy v3.3; effective date 2026-05-26.","versionDate":"2026-05-26","canonicalUrl":"https://cdn.sanity.io/files/4zrzovbb/website/c11e84981d0a7281a1b229f3fa6af0da66eaf43f.pdf","evidenceUrl":"https://cdn.sanity.io/files/4zrzovbb/website/c11e84981d0a7281a1b229f3fa6af0da66eaf43f.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260528170023/https://cdn.sanity.io/files/4zrzovbb/website/c11e84981d0a7281a1b229f3fa6af0da66eaf43f.pdf","archivedAt":"2026-05-28","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v3.3.","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","title":"Responsible Scaling Policy — v3.4","authors":["Anthropic"],"publisher":"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-3-1","title":"Frontier Safety Framework — v3.1","authors":["Google DeepMind"],"publisher":"Google DeepMind","publicationDate":"2026-04-17","version":"Frontier Safety Framework v3.1; publication date 2026-04-17.","versionDate":"2026-04-17","canonicalUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3-1.pdf","evidenceUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3-1.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260423021420/https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3-1.pdf","archivedAt":"2026-04-23","accessedAt":"2026-09-18","reuse":{"attribution":"Google DeepMind, Frontier Safety Framework — v3.1.","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-pf-beta","title":"Preparedness Framework — Beta","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2023-12-18","version":"Preparedness Framework Beta; publication date 2023-12-18.","versionDate":"2023-12-18","canonicalUrl":"https://cdn.openai.com/openai-preparedness-framework-beta.pdf","evidenceUrl":"https://cdn.openai.com/openai-preparedness-framework-beta.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20231218195930/https://cdn.openai.com/openai-preparedness-framework-beta.pdf","archivedAt":"2023-12-18","accessedAt":"2026-09-18","reuse":{"attribution":"OpenAI, Preparedness Framework — Beta.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-fgf-2026","title":"Frontier Governance Framework — May 2026 edition","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2026-05-28","version":"Frontier Governance Framework May 2026 edition; publication date 2026-05-28.","versionDate":"2026-05-28","canonicalUrl":"https://cdn.openai.com/pdf/e37d949b-8c9f-4d76-b99e-4272f4631a7e/openai-frontier-governance-framework.pdf","evidenceUrl":"https://cdn.openai.com/pdf/e37d949b-8c9f-4d76-b99e-4272f4631a7e/openai-frontier-governance-framework.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260529171758/https://cdn.openai.com/pdf/e37d949b-8c9f-4d76-b99e-4272f4631a7e/openai-frontier-governance-framework.pdf","archivedAt":"2026-05-29","accessedAt":"2026-09-18","reuse":{"attribution":"OpenAI, Frontier Governance Framework — May 2026 edition.","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-faif-2026","title":"Frontier Artificial Intelligence Framework — June 2026 edition","authors":["xAI"],"publisher":"xAI","publicationDate":null,"version":"Frontier Artificial Intelligence Framework June 2026 edition; effective date 2026-06-30. Visible public cover identifies the effective edition; stale embedded PDF draft metadata is not used to infer publication status.","versionDate":"2026-06-30","canonicalUrl":"https://media.x.ai/v1/website/xai-frontier-artificial-intelligence-framework-30-june-2026-99c40684.pdf","evidenceUrl":"https://media.x.ai/v1/website/xai-frontier-artificial-intelligence-framework-30-june-2026-99c40684.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260716003429/https://media.x.ai/v1/website/xai-frontier-artificial-intelligence-framework-30-june-2026-99c40684.pdf","archivedAt":"2026-07-16","accessedAt":"2026-09-18","reuse":{"attribution":"xAI, Frontier Artificial Intelligence Framework — June 2026 edition.","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-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-erewhon-1901-gutenberg","title":"Erewhon, or Over the Range (revised edition, with the prefaces of 1872 and 1901)","authors":["Samuel Butler"],"publisher":"Project Gutenberg (e-text 1906)","publicationDate":"1901","version":"Project Gutenberg e-text 1906 (released 1999, updated 2025) of the revised edition; it does not name its source printing. Retrieved 2026-10-06; no printed 1901 copy was compared.","canonicalUrl":"https://www.gutenberg.org/ebooks/1906","evidenceUrl":"https://www.gutenberg.org/cache/epub/1906/pg1906.txt","language":"en","archivedUrl":"https://web.archive.org/web/20260619211126/https://www.gutenberg.org/cache/epub/1906/pg1906.txt","archivedAt":"2026-06-19","accessedAt":"2026-10-06","reuse":{"attribution":"Samuel Butler, Erewhon (revised edition, 1901), Project Gutenberg e-text 1906.","licence":"Public domain (published 1872 and 1901; the author died in 1902).","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each) under Selekoa’s quotation policy. Public-domain text."}},{"id":"source-butler-1865-press","title":"Lucubratio Ebria","authors":["Unsigned (attributed to Samuel Butler)"],"publisher":"The Press (Christchurch)","publicationDate":"1865-07-29","version":"Uncorrected Papers Past OCR text in the Internet Archive capture of 2026-09-12. The column images exist only as HTTP 403 captures, so no page image was compared.","canonicalUrl":"https://paperspast.natlib.govt.nz/newspapers/CHP18650729.2.8","evidenceUrl":"https://web.archive.org/web/20260912064521/https://paperspast.natlib.govt.nz/newspapers/CHP18650729.2.8","language":"en","archivedUrl":"https://web.archive.org/web/20260912064521/https://paperspast.natlib.govt.nz/newspapers/CHP18650729.2.8","archivedAt":"2026-09-12","accessedAt":"2026-10-06","reuse":{"attribution":"The Press (Christchurch), 29 July 1865, via Papers Past (National Library of New Zealand).","licence":"Public domain (published 1865). Papers Past: no known copyright (New Zealand).","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each) under Selekoa’s quotation policy. Public-domain text."}},{"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-hanson-2008-core-argument","title":"What Core Argument?","authors":["Robin Hanson"],"publisher":"Overcoming Bias","publicationDate":"2008-12","version":"Current page on Substack, reached from the 2008 address and retrieved 2026-10-06; its bytes differ from the 1996–2022 candidate search’s retrieval while the quoted text matches.","canonicalUrl":"https://www.overcomingbias.com/p/what-core-argumenthtml","evidenceUrl":"https://www.overcomingbias.com/2008/12/what-core-argument.html","language":"en","archivedUrl":"https://web.archive.org/web/20260810091248/https://www.overcomingbias.com/p/what-core-argumenthtml","archivedAt":"2026-08-10","accessedAt":"2026-10-06","reuse":{"attribution":"Robin Hanson, \"What Core Argument?\", Overcoming Bias, December 2008.","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-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-extropians-1998-09-18-yudkowsky-2487","title":"Re: Singularity: The Moore's Law Argument (Extropians mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"Extropians mailing list","publicationDate":"1998-09-18","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 6b53d7a85ac8fa352ccda20157a393ef5d212919ceae95168353a5a189d1c2ed); quoted lines excluded when reading the author’s text. The served archive shows later URL rewriting and does not authenticate historical bytes. Archive at extropians.weidai.com.","canonicalUrl":"http://extropians.weidai.com/extropians.3Q98/2487.html","evidenceUrl":"http://extropians.weidai.com/extropians.3Q98/2487.html","language":"en","archivedUrl":"https://web.archive.org/web/20250624043401/http://extropians.weidai.com/extropians.3Q98/2487.html","archivedAt":"2025-06-24","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: Singularity: The Moore's Law Argument”, Extropians mailing list, 1998-09-18.","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-2023-gpts-are-predictors-not-imitators","title":"GPTs are Predictors, not Imitators","authors":["Eliezer Yudkowsky"],"publisher":"LessWrong","publicationDate":"2023-04-08","version":"Post body retrieved through the LessWrong GraphQL API on 2026-10-06 (SHA-256 24a41ca1ec6d0276008e11e705db91b8a645ad70bc2fcad1245ca1a597f5528a), content version 1.1.0; quotations also compared with the posting-day version 1.0.0.","canonicalUrl":"https://www.lesswrong.com/posts/nH4c3Q9t9F3nJ7y8W/gpts-are-predictors-not-imitators","evidenceUrl":"https://www.lesswrong.com/posts/nH4c3Q9t9F3nJ7y8W/gpts-are-predictors-not-imitators","language":"en","archivedUrl":"https://web.archive.org/web/20260930171831/https://www.lesswrong.com/posts/nH4c3Q9t9F3nJ7y8W/gpts-are-predictors-not-imitators","archivedAt":"2026-09-30","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, “GPTs are Predictors, not Imitators”, LessWrong, 2023-04-08.","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-sachs-kotlikoff-2012-smart-machines","title":"Smart Machines and Long-Term Misery","authors":["Jeffrey D. Sachs","Laurence J. Kotlikoff"],"publisher":"NBER Working Paper 18629","publicationDate":"2012-12","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 aa4958092702ecfad60c57fcdff3a76505353fccab0bcb60bda2fb1b14c0a65b.","canonicalUrl":"https://www.nber.org/system/files/working_papers/w18629/w18629.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w18629/w18629.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Jeffrey D. Sachs, Laurence J. Kotlikoff, “Smart Machines and Long-Term Misery”, NBER Working Paper 18629, 2012-12.","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-furman-2019-automation-future-past","title":"Should We Be Reassured If Automation in the Future Looks Like Automation in the Past?","authors":["Jason Furman"],"publisher":"NBER volume chapter","publicationDate":"2019-05","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 a43ecf48d5ec6131dfc002acab3d7df32dffb90bb9b7e41c30c470e14e6e68f6.","canonicalUrl":"https://www.nber.org/system/files/chapters/c14031/c14031.pdf","evidenceUrl":"https://www.nber.org/system/files/chapters/c14031/c14031.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Jason Furman, “Should We Be Reassured If Automation in the Future Looks Like Automation in the Past?”, NBER volume chapter, 2019-05.","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-jones-2024-ai-dilemma-aer-insights","title":"The AI Dilemma: Growth versus Existential Risk","authors":["Charles I. Jones"],"publisher":"American Economic Review: Insights 6(4), December 2024","publicationDate":"2024-12","version":"AEA abstract page. Retrieved 2026-10-11. SHA-256 a39825e1555406539235c485b38d1841de5b0604dcfbae59fe443c5143ea48e2. Version a reader would cite, or full text, not retrieved. Jones, C. I. (2024). The AI dilemma: Growth versus existential risk. American Economic Review: Insights, 6(4). https://doi.org/10.1257/aeri.20230570 (full text not retrieved)","canonicalUrl":"https://www.aeaweb.org/articles?id=10.1257/aeri.20230570","evidenceUrl":"https://www.aeaweb.org/articles?id=10.1257/aeri.20230570","language":"en","accessedAt":"2026-10-11","reuse":{"attribution":"Charles I. Jones, “The AI Dilemma: Growth versus Existential Risk”, American Economic Review: Insights 6(4), December 2024, 2024-12.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-cn-zhipu-prospectus-en","title":"Knowledge Atlas Technology Joint Stock Company Limited (北京智譜華章科技股份有限公司) — Global Offering","authors":["Knowledge Atlas Technology Joint Stock Company Limited"],"publisher":"Knowledge Atlas Technology Joint Stock Company Limited (via HKEXnews)","publicationDate":"2025-12-30","version":"Dated document, retrieved on 2026-10-15 (SHA-256 a6598ce3f2f13ac78a9d02a85314dcd349314d2ee1215015e41bb1284e608f4e) directly from the publisher.","canonicalUrl":"https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1230/2025123000017.pdf","evidenceUrl":"https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1230/2025123000017.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20261002230645/https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1230/2025123000017.pdf","archivedAt":"2026-10-02","accessedAt":"2026-10-15","reuse":{"attribution":"Knowledge Atlas Technology Joint Stock Company Limited, “Knowledge Atlas Technology Joint Stock Company Limited (北京智譜華章科技股份有限公司) — Global Offering”, 2025-12-30.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-goo-author-hassabis","title":"Demis Hassabis (author page, undated)","authors":["Google (The Keyword)"],"publisher":"Google","publicationDate":null,"version":"Undated page, retrieved on 2026-10-15 (SHA-256 2fac6eac5f83946b0731f4695199d7e518db71abae11e6e1353f44b0c0f0d07d) directly from the publisher. Its wording can change; earlier wording is not authenticated. Publisher as printed on the inspected expression: Google (The Keyword).","canonicalUrl":"https://blog.google/authors/demis-hassabis/","evidenceUrl":"https://blog.google/authors/demis-hassabis/","language":"en","archivedUrl":"https://web.archive.org/web/20261004210336/https://blog.google/authors/demis-hassabis/","archivedAt":"2026-10-04","accessedAt":"2026-10-15","reuse":{"attribution":"Google (The Keyword), “Demis Hassabis (author page, undated)”.","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-goo-pichai-2026-08","title":"The next chapter of our AI momentum","authors":["Sundar Pichai","Demis Hassabis"],"publisher":"Google","publicationDate":"2026-08-05","version":"Dated page, retrieved on 2026-10-15 (SHA-256 26a44372e5ac41a9a18bb4541b4534f558827fa91d6c9431d136ded14760f76e) directly from the publisher.","canonicalUrl":"https://blog.google/company-news/inside-google/message-ceo/next-chapter-ai-momentum/","evidenceUrl":"https://blog.google/company-news/inside-google/message-ceo/next-chapter-ai-momentum/","language":"en","archivedUrl":"https://web.archive.org/web/20261011055958/https://blog.google/company-news/inside-google/message-ceo/next-chapter-ai-momentum/","archivedAt":"2026-10-11","accessedAt":"2026-10-15","reuse":{"attribution":"Sundar Pichai, Demis Hassabis, “The next chapter of our AI momentum”, 2026-08-05.","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-us-eo-14409-fr","title":"Executive Order 14409 of June 2, 2026: Promoting Advanced Artificial Intelligence Innovation and Security (91 FR 34565)","authors":["The White House"],"publisher":"The White House","publicationDate":"2026-06-05","version":"Retrieved on 2026-10-27 (SHA-256 4a25c970947e7234f1a4308aca6aba62f327ac956de7caa41204c84f7f2cdcd9). Federal Register print of 5 June 2026, Vol. 91, No. 108, pp. 34565–34567, FR Doc. 2026-11415. Publisher as printed on the inspected expression: Federal Register (Office of the Federal Register, National Archives) via govinfo.gov.","canonicalUrl":"https://www.govinfo.gov/content/pkg/FR-2026-06-05/pdf/2026-11415.pdf","evidenceUrl":"https://www.govinfo.gov/content/pkg/FR-2026-06-05/pdf/2026-11415.pdf","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"The White House, “Executive Order 14409 of June 2, 2026: Promoting Advanced Artificial Intelligence Innovation and Security (91 FR 34565)”, 2026-06-05.","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-ab-eo-14292-fr","title":"Executive Order 14292 of May 5, 2025: Improving the Safety and Security of Biological Research (90 FR 19611)","authors":["Federal Register (Office of the Federal Register, National Archives) via govinfo.gov"],"publisher":"Federal Register (Office of the Federal Register, National Archives) via govinfo.gov","publicationDate":null,"version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 a51c01bb2501ba0570461ad40384a15aa482d24cd7b4c4c36923f8e9bbc62e73). Federal Register print of 8 May 2025, Vol. 90, No. 88, pp. 19611-19613, FR Doc. 2025-08266. Retrieved again at admission on 2026-10-30 (SHA-256 a51c01bb2501ba0570461ad40384a15aa482d24cd7b4c4c36923f8e9bbc62e73, the same bytes).","canonicalUrl":"https://www.govinfo.gov/content/pkg/FR-2025-05-08/pdf/2025-08266.pdf","evidenceUrl":"https://www.govinfo.gov/content/pkg/FR-2025-05-08/pdf/2025-08266.pdf","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Federal Register (Office of the Federal Register, National Archives) via govinfo.gov, “Executive Order 14292 of May 5, 2025: Improving the Safety and Security of Biological Research (90 FR 19611)”.","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-fsf-v3-1-cbrn-tcls","sourceRole":"primary","support":"supports","locator":"PDF p. 17, §5.3 Past Updates and Changes, Version 3.1 (April 17, 2026), first bullet","passage":"Introduced CBRN TCLs and outlined mitigation and risk acceptance process.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical to the Internet Archive capture of 2026-04-23; it replaces an earlier attributed paraphrase. Human editorial review is pending."},{"id":"evidence-absent-faif-2026-benefits-may-outweigh-benchmark-risks","sourceRole":"primary","support":"supports","locator":"June 2026 edition, PDF pp. 1–9, whole document: §1 Introduction; §2 Risk Management Framework with §2.1 (a)–(c), §2.2, both sections numbered 2.3 and §2.4; §3 Incident reporting; §4 Governance","passage":"Within the inspected scope, no passage stating the provision “Deployment benefits may outweigh benchmark-identified risks” was found.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-03","note":"Absence recorded by an automated research assistant on 2026-10-03. Method: Read the whole document, in particular both sections numbered 2.3 (Systemic risk acceptance determination; Safety mitigations); searched the text layer for “outweigh”, “conceivable”, “benefit” and “balance”. No passage stating that the expected benefits of deployment may outweigh the risks identified by a particular benchmark was found within PDF pp. 1–9. Text comparison only. No publisher statement establishes that the June 2026 edition replaces the December 2025 edition. Human editorial review is pending."},{"id":"evidence-prov-rsp-v3-4-ltbt-review-request","sourceRole":"primary","support":"supports","locator":"PDF p. 14 (printed p. 14), §3.6 External Review, final paragraph before §3.6.1","passage":"In addition, upon the LTBT’s request, we will conduct a public or private external review of our Risk Report or sections thereof, subject to the procedures outlined below.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical to the Internet Archive capture of 2026-07-10. Extraction note: Wording identical to v3.3; no change proposed. Human editorial review is pending."},{"id":"evidence-provchange-rsp-v3-4-automated-rd-threshold-later","sourceRole":"primary","support":"supports","locator":"PDF pp. 9–10 (printed pp. 9–10), §1 table row 'Automated R&D in key domains', left column, paragraph beginning 'We would consider scenario (2)', with footnote 9","passage":"v3.4 measures the doubling of the rate of progress in AI aggregate capabilities against both the rate Anthropic would expect and the fastest rate of extended progress (footnote 9: over at least three model generations) it has observed without significant AI contributions to AI R&D, and asks whether continuing the trend seems likely to lead to even greater acceleration.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-03","note":"Paraphrase checked by an automated research assistant on 2026-10-03 against the cited edition; the edition’s own wording is quoted in its provision record. Human editorial review is pending."},{"id":"evidence-provchange-rsp-v3-4-automated-rd-threshold-earlier","sourceRole":"primary","support":"supports","locator":"PDF pp. 9–10 (printed pp. 9–10), §1 table row 'Automated R&D in key domains', left column, paragraph beginning 'We would consider scenario (2)'","passage":"v3.3 measures the doubling of the rate of progress in AI aggregate capabilities against the rate Anthropic would expect without significant AI contributions to AI R&D, and asks whether continuing the trend could lead to even greater acceleration.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-03","note":"Paraphrase checked by an automated research assistant on 2026-10-03 against the cited edition; the edition’s own wording is quoted in its provision record. Human editorial review is pending."},{"id":"evidence-provchange-rsp-v3-4-automated-rd-threshold-changelog","sourceRole":"primary","support":"supports","locator":"Changelog entry 'July 8, 2026 (RSP v3.4)', item 1, PDF p. 20 (printed p. 20)","passage":"The v3.4 change log says, as item 1, that the update revises the Automated R&D capability threshold in light of further discussion, particularly whether the threshold would count as crossed if the overall rate of AI progress were constant or slowing but estimated to be dramatically faster than without advanced AI tools; Anthropic says it would not, and that the threshold is intended to capture the onset of dramatic recursive self-improvement.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-03","note":"Paraphrase checked by an automated research assistant on 2026-10-03 against the cited edition; the edition’s own wording is quoted in its provision record. Human editorial review is pending."},{"id":"evidence-ts-b6-butler-1865-contests","sourceRole":"primary","support":"supports","locator":"Paragraph beginning \"It is a mistake then to take the view\", after the comparison with the Chinese and with \"savage tribes\"","passage":"It is a mistake then to take the view adopted by a previous correspondent [...] to consider the machines as identities, [...] and to anticipate their final triumph over mankind. They are to be regarded as the mode of development by which human organism is most especially advancing [...]","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Uncorrected OCR in the 2026-09-12 capture; the elided words include the OCR error \"aniinalise\". No page image was available, so the wording is not image-verified. Re-retrieved 2026-10-06; the capture matched the hash recorded by the pre-1996 candidate search."},{"id":"evidence-ts-b6-butler-1865-contests-2","sourceRole":"primary","support":"context","locator":"Preface to the Revised Edition (dated 1901-08-07), paragraph after the account of the 1863 letter","passage":"A second article on the same subject as the one just referred to appeared in the Press shortly after the first, but I have no copy.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Butler’s account of a second Press article on machines; it is given no title or date, and the two-year interval does not fit \"shortly after\". Context for attribution only."},{"id":"evidence-ts-a5-good-1965","sourceRole":"primary","support":"supports","locator":"Section 2, printed p. 33 (scan page 3)","passage":"Because \"the design of machines is one of these intellectual activities\", an ultraintelligent machine could design even better machines.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-06","note":"Checked against the text layer of the inspected scan on 2026-10-06."},{"id":"evidence-ts-a4-vinge-1993","sourceRole":"primary","support":"supports","locator":"Printed p. 14 (PDF page 4), final paragraphs before \"Can the Singularity be Avoided?\"","passage":"And what happens a month or two (or a day or two) after that? I have only analogies to point to: The rise of humankind. We will be in the Post-Human era.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Checked against the text layer on 2026-10-06."},{"id":"evidence-ts-a4-vinge-1993-2","sourceRole":"primary","support":"supports","locator":"Printed p. 14 (PDF page 4), section \"What is The Singularity?\"","passage":"Since it involves an intellectual runaway, it will probably occur faster than any technical revolution seen so far.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Checked against the text layer on 2026-10-06; the sentence alone, without the period above, would be context."},{"id":"evidence-ts-a2-hanson-2008-core-argument-contests","sourceRole":"primary","support":"supports","locator":"First paragraph; chapter 53 of the MIRI collection, printed p. 397","passage":"a single AI to within roughly a week grow from unimportant to strong enough to take over the world. [...] This amazing growth rate [...] is far beyond what the vast majority of AI researchers, growth economists, or most any other specialists would estimate.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Checked against the current page on 2026-10-06."},{"id":"evidence-yk-t10-e3","sourceRole":"primary","support":"supports","locator":"message body, paragraph beginning 'The scenario above points out'","passage":"Once intelligence can be enhanced by technology, the rate of progress in technology is a function of how far you've already gone.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-07","note":"Candidate pack passage T10-e3. The quotation (22 words) matched the retrieved expression (SHA-256 6b53d7a85ac8fa352ccda20157a393ef5d212919ceae95168353a5a189d1c2ed) when checked on 2026-10-07."},{"id":"evidence-yk-t21-a","sourceRole":"primary","support":"supports","locator":"GPTs are Predictors, not Imitators, closing paragraphs","passage":"since the task that GPTs are being trained on is different from and harder than the task of being a human, it would be surprising … if GPTs ended up thinking the way humans do, in order to solve that problem.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T21-a. The quotation (41 words) matched the retrieved expression (SHA-256 24a41ca1ec6d0276008e11e705db91b8a645ad70bc2fcad1245ca1a597f5528a) when checked on 2026-10-06. It also matched the posting-day version 1.0.0. Pack version note: edited post, current version 1.1.0; posting-day check pending."},{"id":"evidence-yk-t26-c","sourceRole":"primary","support":"supports","locator":"AGI Ruin: A List of Lethalities, Section A, item 2, first sentence","passage":"A cognitive system with sufficiently high cognitive powers, given any medium-bandwidth channel of causal influence, will not find it difficult to bootstrap to overpowering capabilities","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T26-c. The quotation (25 words; trimmed 2026-10-13 from 29 words) matched the retrieved expression (SHA-256 b36f38a6fd1020bfbd7b9d941e150ea68561242f3a41a0b8f2a56b853a9a6db3) when checked on 2026-10-06. It also matched the posting-day version 1.0.0. Pack version note: edited post, current version 1.3.0; posting-day check pending."},{"id":"evidence-ec-m03-09","sourceRole":"primary","support":"supports","locator":"pdf page 5; 12.1 The Benefit of More Artificial Intelligence (printed p. 320)","passage":"However, while AI research has been underway for decades, recent advances are still very new, and, as a result, AI has not had a large macroeconomic impact, at least not yet.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M03-09 (research/economics-mainstream). The quotation (31 words) matched the retrieved current copy (SHA-256 a43ecf48d5ec6131dfc002acab3d7df32dffb90bb9b7e41c30c470e14e6e68f6) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m06b-09","sourceRole":"primary","support":"supports","locator":"pdf page 18; Conclusion (printed p. 17)","passage":"When the older generation enjoys a windfall from the advance of technology, the government can tax some of that windfall, and then use the proceeds to improve the wellbeing of today's youth and of future generations.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M06b-09 (research/economics-mainstream). The quotation (36 words) matched the retrieved current copy (SHA-256 aa4958092702ecfad60c57fcdff3a76505353fccab0bcb60bda2fb1b14c0a65b) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-e03-16","sourceRole":"primary","support":"supports","locator":"Abstract (publisher abstract page)","passage":"This paper considers the optimal use of AI technology in the presence of these opportunities and risks. Under what conditions should we continue the rapid progress of AI, and under what conditions should we stop?","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-16","note":"Candidate pack passage E03-16 (research/economics-mainstream/esoteric). The quotation (35 words) matched the retrieved current copy (SHA-256 a39825e1555406539235c485b38d1841de5b0604dcfbae59fe443c5143ea48e2) in the pack's verification of 2026-10-15. Re-checked at admission on 2026-10-16 against a copy retrieved that day, whose SHA-256 differs from the recorded one.","contextOnly":{"reason":"Fresh-context audit (A1) judged that the quotation misleads when shown alone.","requires":"The quote's 'these opportunities and risks' refers back to the preceding abstract sentence ('many experts worry ... existential risk: creating a superintelligence misaligned with human values could lead to ... human extinction'); the quote alone never names existential risk. The risk premise is attributed to 'many experts', and the optimisation is a model, not a forecast."}},{"id":"lg-ev-cn-wang-shaolan-zhipu-deputy-gm-2025","sourceRole":"primary","support":"supports","locator":"Directors, Supervisor and Senior Management, “Senior Management”, printed page 230","passage":"is our co-founder and deputy general manager. He is mainly responsible for the business development, financing","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":"lg-ev-goo-hassabis-alphabet-2026","sourceRole":"primary","support":"supports","locator":"Sundar's message, AGI and science paragraph","passage":"Demis will become the Chair of GDM and Chief Scientist of Alphabet","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (12 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-goo-hassabis-alphabet-author-page","sourceRole":"primary","support":"context","locator":"Author title","passage":"The undated author page, as retrieved on 15 October 2026, lists him as Chair, Google DeepMind and Chief Scientist, Alphabet; it shows a current title, not the announcement.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “Chair, Google DeepMind and Chief Scientist, Alphabet” 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":"ev-cmp-fsf-v3-1-d3-halt-1","sourceRole":"primary","support":"supports","locator":"PDF p. 10 (printed p. 10), §2.1.2 Deployment Mitigations, numbered step 2, first sentence","passage":"Pre-deployment review of residual risk assessment: external deployments of a model take place only after the appropriate governance function determines the residual risk to be acceptable (including a safety case where a CCL has been reached).","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Existing dataset passage for this edition, which contains the comparison’s shorter quotation, first checked on 2026-10-03. The comparison’s quotation was 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. The sentence continues '(including a safety case where a CCL has been reached)'. Human editorial review is pending."},{"id":"ev-cmp-faif-2026-d6-discretion-1","sourceRole":"primary","support":"supports","locator":"PDF p. 6, first §2.3 'Systemic risk acceptance determination', third paragraph, third sentence","passage":"The final decision also reflects an overall judgment based on all the available evidence","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-15 against the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. Human editorial review is pending."},{"id":"ev-cmp-faif-2026-d6-discretion-2","sourceRole":"primary","support":"supports","locator":"PDF p. 7, second §2.3 'Safety mitigations', paragraph following the safeguards list, second sentence","passage":"xAI may change its approach from that listed above in order to make additional improvements.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-15 against the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. Human editorial review is pending."},{"id":"ev-cmp-meta-aasf-v2-0-d6-competitor","sourceRole":"primary","support":"supports","locator":"Meta AASF · v2: Whole text layer, PDF pp. 1–44 (p. 45 blank): all sections, tables, footnotes and both appendices read","passage":"Within the inspected scope, no passage meeting the comparison question “Provisions conditioned on other developers” was found.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-26","note":"Absence recorded by an automated research assistant on 2026-10-15 for the framework comparison, audited in a fresh context, checked by a reviewer from a different model family on 2026-10-15 and re-checked on 2026-10-26. Human editorial review is pending."},{"id":"ev-cmp-preparedness-beta-2023-d1-cyber-1","sourceRole":"primary","support":"supports","locator":"PDF p. 8, Tracked Risk Categories › 'Cybersecurity', opening sentence","passage":"The edition describes its Cybersecurity category as concerned with risks of the model being used for cyber-exploitation that disrupts the confidentiality, integrity or availability of computer systems.","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":"ev-cmp-frontier-governance-2026-d7-internal-1","sourceRole":"primary","support":"supports","locator":"PDF p. 13 (printed p. 11), §2.5 Safety mitigations, second paragraph, second sentence","passage":"The edition says OpenAI documents a justification of why a model’s systemic risks are acceptable, which includes the reasonably foreseeable conditions under which that justification may stop holding, informed by recommendations from its Safety Advisory Group and outside experts, “as available and appropriate”.","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 it attached the qualifier “as available and appropriate” to the external experts alone, where the passage places it after the whole clause; the correction keeps the passage’s ambiguity and does not resolve it. 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 passage does not say whether “as available and appropriate” qualifies the Safety Advisory Group’s recommendations, the external experts’ or both; the paraphrase leaves that open. Human editorial review is pending."},{"id":"evidence-ab-scr-eo14292-signed-1","sourceRole":"primary","support":"supports","locator":"Heading, 90 FR 19611","passage":"Executive Order 14292 of May 5, 2025","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (7 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. The order number and date are taken from the Federal Register print, not from any press report. A cross-family machine check (2026-10-30, item 34) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-gov-us-eo14409-no-biological-provision-1","sourceRole":"primary","support":"supports","locator":"Whole cached text (91 FR 34565-34567)","passage":"the search for the seven terms returned no match; the order's risk vocabulary is cyber (cyber defence, software vulnerabilities, 'advanced cyber capabilities of AI models'), and Section 3(a) ties the 'covered frontier model' designation to a classified benchmarking process for advanced cyber capabilities.","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. The dataset records the order's signature and its Section 3 provisions (recorded separately). A cross-family machine check (2026-10-30, item 92) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"}],"at":{"entities":[18,24,48,49,51,52,53,55,57,187,190,200,232,255,280,318,333,337,339,340,346,359,370,374,398,435,515,519,536,552,587,606,622,1044,1214],"claims":[97,498,546,590,759,790,838,930,1056,1129,1156,1346,1424,1528,1822,1896,2115,2149,2156,2637,2667,3316,3352],"sources":[11,42,43,47,48,50,52,180,254,255,282,323,338,365,402,487,505,555,587,628,643,881,1048],"evidence":[114,529,613,658,659,660,962,963,1000,1049,1050,1152,1285,1371,1404,1678,1771,1907,2341,2449,2450,2743,2788,2789,2800,3311,3345,4551,4602]},"version":"1lg47t7"}