{"entities":[{"id":"openai","name":"OpenAI"},{"id":"google","name":"Google"},{"id":"rsp-v2-1","name":"Anthropic Responsible Scaling Policy — v2.1"},{"id":"rsp-v3-3","name":"Anthropic Responsible Scaling Policy — v3.3"},{"id":"preparedness-beta-2023","name":"OpenAI Preparedness Framework — Beta"},{"id":"preparedness-v2","name":"OpenAI Preparedness Framework — Version 2"},{"id":"faif-2026","name":"xAI Frontier Artificial Intelligence Framework — June 2026 edition"},{"id":"concrete-problems-paper","name":"Concrete Problems in AI Safety"},{"id":"scalable-oversight","name":"Scalable oversight"},{"id":"metr","name":"Model Evaluation & Threat Research"},{"id":"metr-o1-preview-report","name":"METR preliminary o1-preview evaluation (2024)"},{"id":"reinforcement-learning","name":"Reinforcement learning"},{"id":"alphago-zero-account","name":"AlphaGo Zero: Starting from scratch (2017)"},{"id":"dsit","name":"Department for Science, Innovation and Technology"},{"id":"minimax","name":"MiniMax"},{"id":"rlo-2019-deceptive-alignment","name":"Deceptive Alignment (Risks from Learned Optimization, part 4)"},{"id":"switch-off-if-dangerous","name":"Switching it off if it becomes dangerous"},{"id":"tool-without-goals","name":"A tool that wants nothing"},{"id":"extropians-1997-09-26-yudkowsky-4420","name":"Re: Singularity: Generation gap"},{"id":"sl4-2001-02-25-yudkowsky-0102-0594","name":"Re: Military in or out?"},{"id":"eu-ai-act-2024","name":"Regulation (EU) 2024/1689 (Artificial Intelligence Act), as published in the Official Journal"},{"id":"ai-general-purpose-technology-with-lag","name":"AI as a general-purpose technology with lagging productivity"},{"id":"eloundou-et-al-2023-gpts-are-gpts","name":"GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models"},{"id":"hanson-2008-friendly-projects-vs-products","name":"Friendly Projects vs. Products"},{"id":"openai-mission-alignment-team","name":"Mission Alignment team (OpenAI)"},{"id":"scott-garrabrant","name":"Scott Garrabrant"},{"id":"robin-hanson","name":"Robin Hanson"},{"id":"pamela-mishkin","name":"Pamela Mishkin"},{"id":"minimax-m1-report","name":"MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention"},{"id":"bletchley-chair-safety-testing-2023","name":"Safety Testing: Chair's Statement of Session Outcomes, 2 November 2023"},{"id":"uk-google-deepmind-mou-2025","name":"Memorandum of Understanding between the UK and Google DeepMind on AI opportunities and security"},{"id":"us-eo-14110-2023","name":"Executive Order 14110: Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence"},{"id":"us-bureau-of-industry-and-security","name":"Bureau of Industry and Security, U.S. Department of Commerce"},{"id":"us-bis-uae-rule-2026","name":"Enhanced Favorable Treatment for the United Arab Emirates Under the Export Administration Regulations (final rule)"},{"id":"shelby-county-health-department","name":"Shelby County Health Department"},{"id":"bwc-msp-2025-wp28","name":"Statement by Japan (BWC/MSP/2025/WP.28)"}],"claims":[{"id":"lineage-ii-concrete-discusses-oversight","statement":"Concrete Problems in AI Safety discusses scalable oversight when detailed evaluation is costly.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Discussion of one formulation; not earliest terminology or equivalence with later superhuman-evaluation proposals.","evidenceIds":["evidence-lineage-ii-concrete-discusses-oversight"]},{"id":"eval-i-metr-o1-report","statement":"METR published its preliminary o1-preview evaluation report.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"o1-preview access began September 3 and evaluation ended September 9, 2024. Short access, latency and rate limits constrained elicitation; no confident upper bound was established.","evidenceIds":["evidence-eval-i-metr-o1-report"]},{"id":"atlas-alphago-rl","statement":"The AlphaGo Zero account describes reinforcement learning through repeated self-play and search.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Authors’ description, not an independently reproduced result or a claim that Go methods transfer to biology.","evidenceIds":["evidence-atlas-alphago-rl"]},{"id":"prov-rsp-v3-3-marginal-risk-board-ltbt-approval","statement":"RSP v3.3 states that, where marginal risk analysis plays a major role in a decision to move forward, explicit approval of the Risk Report by the Board and LTBT, rather than just the CEO and RSO, will be required.","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-v3-3-marginal-risk-board-ltbt-approval"],"attributes":{"provision":{"kind":"decision_authority","key":"rsp-marginal-risk-board-ltbt-approval","parties":[{"role":"decides","actor":"Board of Directors"},{"role":"decides","actor":"Long-Term Benefit Trust"},{"role":"decides","actor":"CEO"},{"role":"decides","actor":"Responsible Scaling Officer"}]}}},{"id":"prov-rsp-v3-3-automated-rd-threshold","statement":"RSP v3.3 states that it would consider the dramatic-acceleration scenario of the Automated R&D threshold to have occurred where, among other conditions, Anthropic observes or expects double the rate of progress in AI aggregate capabilities compared to the rate it would expect in the absence of significant AI contributions to AI R&D.","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-v3-3-automated-rd-threshold"],"attributes":{"provision":{"kind":"threshold","key":"rsp-automated-rd-threshold"}}},{"id":"provchange-preparedness-v2-final-decision-authority","statement":"Version 2’s passage states that OpenAI Leadership can approve or reject the SAG’s recommendations and that the Board’s Safety and Security Committee provides oversight of these decisions; the Beta’s passage names OpenAI Leadership as final decision-maker with the option for the OpenAI Board of Directors to overrule.","derivation":"inferred","resolution":"supported","inferenceRationale":"Premises: Preparedness · Version 2 states the provision as quoted in its provision record (PDF p. 4 (printed p. 3), §1 Introduction, final paragraph, second sentence); Preparedness · Beta states it as quoted in its provision record (PDF p. 22 (printed p. 22), Governance › “Operations”, opening paragraph, second sentence, items (2) and (3)). The difference described is read directly from the two passages.","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. Comparison only; it does not assert that Version 2 replaces the Beta.","evidenceIds":["evidence-provchange-preparedness-v2-final-decision-authority-later","evidence-provchange-preparedness-v2-final-decision-authority-earlier"],"attributes":{"changeBasis":"document_comparison","provision":{"kind":"decision_authority","key":"pf-final-decision-authority"}}},{"id":"yk-y01-extropians-1997-09-26-yudkowsky-4420-c","statement":"The Extropians message “Re: Singularity: Generation gap” (1997-09-26) contests Y01. In a reply to Eric Watt Forste about Singularity attitudes, he reports that a correspondent proposed simply pulling the plug on a transhuman intelligence if it looked dangerous, and answers that such a precaution would fail even against a human-level mind that had copied itself across the Internet.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: A capable goal-directed system will not act to prevent its own shutdown or modification, and a shutdown or correction mechanism can simply be added to it. The second half is ironic role-play in the voice of the escaping intelligence. Earliest located direct answer to the 'pull the plug' proposal; the earliest passage in the answer/contrary group is his Extropians message of 1996-12-08. The same paragraph also dismisses 'keep it bottled up … only if it convinces us' (‘Confinement without network access keeps it safe’, Y06), but that is later than his messages of 1996-12-31 recorded for it.","evidenceIds":["evidence-yk-t01-e2"],"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 a capable goal-directed system will not act to prevent its own shutdown or modification, and a shutdown or correction mechanism can simply be added to it."}}}},{"id":"yk-y15-sl4-2001-02-25-yudkowsky-0102-0594-c","statement":"The SL4 message “Re: Military in or out?” (2001-02-25) contests Y15. Replying to Ben Goertzel on military AI, he describes 'toollike' superintelligences structured around immediate instructions rather than supergoals (which could be told to kill every living thing in a country) and answers his own question whether such AI can be built safely.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: A system that only outputs answers or plans, leaving humans to act on them, is safe. The Q/A is his own rhetorical device, not quoted from anyone. A bare conclusion; the argument is only implied by the earlier paragraph about instructing a toollike SI. Earlier than his post ‘The Rhythm of Disagreement’ (2008-06-01).","evidenceIds":["evidence-yk-t15-s1"],"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 system that only outputs answers or plans, leaving humans to act on them, is safe, even when read with its message or dialogue turn."}}}},{"id":"ec-m03-eloundou-et-al-2023-gpts-are-gpts-b","statement":"Eloundou et al., “GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models” (2023-03-17) bears on M03. The authors argue LLMs meet general-purpose-technology criteria of pervasiveness, improvement and complementary innovation. The authors note that co-invention requirements typical of general-purpose technologies may slow LLM diffusion.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"(First passage) Wording reproduced as in arXiv v1 (including 'suggesting that as'). (Second passage) Diffusion lag due to co-invention, stated as possibility.","evidenceIds":["evidence-ec-m03-10","evidence-ec-m03-11"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"(M03-10) No statement that measured productivity gains lag adoption. (M03-11) Refers to slowed diffusion, not to measured productivity lagging adoption.","tier":{"value":1,"basis":"Scholarly working paper with an economist author: Daniel Rock (D2, 2026-10-11, confirmed); the other three were at OpenAI or OpenResearch, as the paper states."}}}},{"id":"lg-gs-mission-alignment-team-disbanded-2026","statement":"TechCrunch reported on 11 February 2026 that OpenAI had disbanded its mission alignment team and had confirmed that the team's members had been assigned to other roles.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The record rests on reporting (TechCrunch), with OpenAI's confirmation relayed by TechCrunch; a spokesperson told TechCrunch that the project's work continues throughout the organisation. The record documents a team ended as a unit and its members reassigned within OpenAI; the record does not establish where each member went.","evidenceIds":["lg-ev-gs-mission-alignment-team-disbanded-2026-a","lg-ev-gs-mission-alignment-team-disbanded-2026"]},{"id":"ap-scott-garrabrant-rlo-2019-deceptive-alignment","statement":"“Deceptive Alignment (Risks from Learned Optimization, part 4)” (2019-06-05) credits Scott Garrabrant as one of its 5 authors.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Credited co-authorship; it does not allocate parts of the text between authors or imply agreement with every passage.","evidenceIds":["evidence-ap-scott-garrabrant-rlo-2019-deceptive-alignment"]},{"id":"ap-robin-hanson-hanson-2008-friendly-projects-vs-products","statement":"“Friendly Projects vs. Products” (2013) credits Robin Hanson 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-robin-hanson-hanson-2008-friendly-projects-vs-products"]},{"id":"ap-pamela-mishkin-eloundou-et-al-2023-gpts-are-gpts","statement":"“GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models” (2023-03-17) credits Pamela Mishkin as one of its 4 authors.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Credited co-authorship; it does not allocate parts of the text between authors or imply agreement with every passage.","evidenceIds":["evidence-ap-pamela-mishkin-eloundou-et-al-2023-gpts-are-gpts"]},{"id":"cmp-rsp-v2-1-d1-cyber","statement":"RSP · v2.1 states a provision on the comparison question “Cyber operations”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Listed under “Ongoing Assessment” as a capability that may call for safeguards stronger than ASL-2; the edition sets no Capability Threshold or Required Safeguards for it and says its ongoing assessments aim to determine them in a later iteration of the policy. What the edition states, paraphrased from the cited expression with its page. It is not evidence of implementation, compliance or effectiveness, and a shared question does not make two editions’ provisions equivalent.","evidenceIds":["ev-cmp-rsp-v2-1-d1-cyber-1"],"attributes":{"provision":{"kind":"risk_domain","key":"comparison-d1-cyber","facet":{"id":"D1.cyber","status":"stated"}}}},{"id":"cmp-faif-2026-d2-definition","statement":"FAIF · June 2026 edition states a provision on the comparison question “How a threshold is defined”, 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 edition defines acceptability of systemic risk rather than a capability threshold; the acceptance criteria themselves are not stated. The edition says residual risk takes account of the scale and probability of harm (PDF p. 6). 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-faif-2026-d2-definition-1"],"attributes":{"provision":{"kind":"threshold","key":"comparison-d2-definition","facet":{"id":"D2.definition","status":"partly"}}}},{"id":"c2-minimax-m1-report","statement":"MiniMax posted the MiniMax-M1 technical report to arXiv on 16 June 2025 (version 1); the abstract states that MiniMax publicly releases MiniMax-M1 in versions with 40K and 80K thinking budgets.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The abstract’s ranking claims are out of scope and are not adopted here.","evidenceIds":["c2-ev-minimax-m1-report","c2-ev-minimax-m1-report-2","c2-ev-minimax-m1-report-pros"]},{"id":"fl-intl-bletchley-chair-testing-published","statement":"The chair's statement of session outcomes on safety testing from the AI Safety Summit was published on GOV.UK on 2 November 2023 by the Prime Minister's Office, the Foreign, Commonwealth & Development Office and the Department for Science, Innovation and Technology.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A chair's statement issued by the UK as summit chair; it is not presented as a text signed by the participants.","evidenceIds":["evidence-fl-intl-bletchley-chair-testing-published-1"]},{"id":"fl-uk-gdm-mou-published","statement":"The Department for Science, Innovation and Technology published on GOV.UK on 11 December 2025 a Memorandum of Understanding with Google DeepMind on AI opportunities and security, naming Liz Kendall and Demis Hassabis as signatories without a signing date.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The text names Google LLC as the contracting party ('Google DeepMind'); no signing date is shown, so the effective date (date of the last signature) is unknown.","evidenceIds":["evidence-fl-uk-gdm-mou-published-1","evidence-fl-uk-gdm-mou-published-2"]},{"id":"cpo-ch-bis-uae-2026-published","statement":"The Bureau of Industry and Security published the final rule \"Enhanced Favorable Treatment for the United Arab Emirates Under the Export Administration Regulations\" in the Federal Register of 14 July 2026, effective 10 July 2026.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Records publication and stated effective date. The rule’s defence-trade and non-computing provisions are not recorded.","evidenceIds":["evidence-cpo-ch-bis-uae-2026-published-1","evidence-cpo-ch-bis-uae-2026-published-2"]},{"id":"cpo-dc-google-texas-40bn","statement":"Google announced on 14 November 2025 a $40 billion investment in Texas through 2027 to build cloud and AI infrastructure, including new data center campuses in Armstrong and Haskell Counties.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A stated investment plan for a state, not for one site; no capacity is stated for the campuses.","evidenceIds":["evidence-cpo-dc-google-texas-40bn-1","evidence-cpo-dc-google-texas-40bn-2"],"attributes":{"transactionType":"infrastructure_spend","relationshipContext":"external","transactionStage":"planned","amount":{"value":40000000000,"currency":"USD","bound":"exact","scope":"programme","scopeNote":"Texas, through 2027, as stated"},"statedDates":[{"kind":"target_completion","date":"2027","statedBy":"google"}]}},{"id":"cpo-pw-shelby-xai-turbine-permit","statement":"The Shelby County Health Department issued construction permit 01156-01PC on 2 July 2025 to CTC Property LLC for fifteen Solar SMT-130 natural gas-fired combustion turbines, each rated at 16.48 MW, at xAI’s Memphis facility, with a printed expiry of 2 January 2027, as the cross-family check read the permit. Action News 5 reported the approval on 2 July 2025.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The permit rests on the cross-family check’s reading of an Internet Archive capture of 3 July 2025 (page 1 of the scanned permit, inspected visually); this admission could not reach the Archive, and the county’s own address now answers HTTP 404. The ratings are each turbine’s and are not summed. A construction permit is not a finding on emissions or operation, and the printed expiry is not a statement of current validity. The press report also says the temporary turbines were to be removed by the end of July, and quotes xAI’s welcome and opponents’ criticism; later legal action is not recorded. The report’s “on Thursday” does not match 2 July 2025, a Wednesday; the permit dates the decision.","evidenceIds":["evidence-cpo-pw-shelby-xai-turbine-permit-1","evidence-cpo-pw-shelby-xai-turbine-permit-2","evidence-cpo-pw-shelby-xai-turbine-permit-3","evidence-cpo-pw-shelby-xai-turbine-permit-4","evidence-cpo-pw-shelby-xai-turbine-permit-5"],"attributes":{"quantity":{"value":15,"unit":"turbines","bound":"exact"},"power":{"technology":"gas","mechanism":"onsite_generation"}}},{"id":"ab-gov-us-eo14110-sec44aii-academies-study","statement":"Section 4.4(a)(ii) of Executive Order 14110 directs the Secretary of Defense, within 120 days, to contract with the National Academies of Sciences, Engineering, and Medicine for a study that assesses the ways in which AI can increase biosecurity risks, including risks from generative AI models trained on biological data, and that makes recommendations on mitigating them.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A directive to enter into a contract for a study. It is not evidence that the contract was made or that any study was delivered, and it states no finding about AI and biosecurity. The subsection's further elements, including the national security implications of government-held datasets, are summarised only in the statement's first clause and are not separately recorded. Only the policy-level scope of the commissioned study is cited; the subsection's references to particular data types were not quoted.","evidenceIds":["evidence-ab-gov-us-eo14110-sec44aii-academies-study-1"],"attributes":{"provision":{"kind":"required_action","key":"us-eo14110-sec44aii-national-academies-study"}}},{"id":"ab-gov-eu-ai-act-annex13-biological-sequences","statement":"Point (d) of Annex XIII to Regulation (EU) 2024/1689, which gives the criteria for the designation of general-purpose AI models with systemic risk referred to in Article 51 and which the Commission takes into account for the purpose of determining that a model has capabilities or an impact equivalent to those set out in Article 51(1), point (a), names the model's input and output modalities, the state of the art thresholds for determining high-impact capabilities for each modality, and the specific type of inputs and outputs, giving biological sequences as the example of that type.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Adds the biological example in point (d) to the dataset's record of the Annex XIII criteria. It sets no threshold for biological-sequence models and does not say that any model falls within the Annex; whether any provider is covered is legal application and is not recorded. The Annex is a list of criteria for a designation decision, not a provision addressed to providers.","evidenceIds":["evidence-ab-gov-eu-ai-act-annex13-biological-sequences-1","evidence-ab-gov-eu-ai-act-annex13-biological-sequences-2"],"attributes":{"provision":{"kind":"threshold","key":"eu-ai-act-annex13d-input-output-modalities"}}},{"id":"ab-gov-bwc-japan-wp28-misuse-risk","statement":"The statement by Japan issued as document BWC/MSP/2025/WP.28 at the 2025 Meeting of States Parties states that new technologies such as genome editing technology, synthetic biology and artificial intelligence offer great benefits to humanity but, if misused, could lead to the production of biological weapons that cause devastating harm, and that there are growing concerns that these technologies are becoming available to a wider range of actors.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A statement of one State Party's position, in its own conditional wording. It names no system, no actor and no capability level, and it is not a finding that any technology has been misused.","evidenceIds":["evidence-ab-gov-bwc-japan-wp28-misuse-risk-1"],"attributes":{"provision":{"kind":"risk_domain","key":"bwc-japan-statement-emerging-technology-misuse"}}}],"sources":[{"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-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-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-pf-2","title":"Preparedness Framework — Version 2","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2025-04-15","version":"Preparedness Framework Version 2; revision date 2025-04-15.","versionDate":"2025-04-15","canonicalUrl":"https://cdn.openai.com/pdf/18a02b5d-6b67-4cec-ab64-68cdfbddebcd/preparedness-framework-v2.pdf","evidenceUrl":"https://cdn.openai.com/pdf/18a02b5d-6b67-4cec-ab64-68cdfbddebcd/preparedness-framework-v2.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20250415191354/https://cdn.openai.com/pdf/18a02b5d-6b67-4cec-ab64-68cdfbddebcd/preparedness-framework-v2.pdf","archivedAt":"2025-04-15","accessedAt":"2026-09-18","reuse":{"attribution":"OpenAI, Preparedness Framework — Version 2.","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-concrete-problems-paper-v1","title":"Concrete Problems in AI Safety","authors":["Dario Amodei","Chris Olah","Jacob Steinhardt","Paul Christiano","John Schulman","Dan Mané"],"publisher":"arXiv","publicationDate":"2016-06-21","version":"arXiv:1606.06565v1; first-submission date from version history. Later revisions are not substituted.","versionDate":"2016-06-21","canonicalUrl":"https://arxiv.org/abs/1606.06565","evidenceUrl":"https://arxiv.org/pdf/1606.06565v1","language":"en","accessedAt":"2026-09-25","reuse":{"attribution":"Dario Amodei; Chris Olah; Jacob Steinhardt; Paul Christiano; John Schulman; Dan Mané, Concrete Problems in AI Safety.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; source publications and personal contact details are not redistributed."}},{"id":"source-eval-i-o1-preview","title":"Details about METR’s preliminary evaluation of OpenAI o1-preview","authors":["METR"],"publisher":"METR","publicationDate":"2024-09-12","version":"Dated preliminary report, current migrated web expression inspected 2026-09-27.","canonicalUrl":"https://metr.org/evaluations/openai-o1-preview-report/","evidenceUrl":"https://metr.org/evaluations/openai-o1-preview-report/","language":"en","archivedUrl":"https://web.archive.org/web/20260914201621/https://metr.org/evaluations/openai-o1-preview-report/","archivedAt":"2026-09-14","accessedAt":"2026-09-27","reuse":{"attribution":"METR, Details about METR’s preliminary evaluation of OpenAI o1-preview.","licence":"No open redistribution licence asserted","note":"Attributed paraphrases and bibliographic metadata only; original documents linked, not redistributed."}},{"id":"source-atlas-alphago-zero","title":"AlphaGo Zero: Starting from scratch","authors":["David Silver","Demis Hassabis"],"publisher":"DeepMind","publicationDate":"2017-10-18","version":"Dated publisher page inspected 2026-09-28; current HTML, not an authenticated historical capture.","canonicalUrl":"https://deepmind.google/blog/alphago-zero-starting-from-scratch/","evidenceUrl":"https://deepmind.google/blog/alphago-zero-starting-from-scratch/","language":"en","archivedUrl":"https://web.archive.org/web/20260924104711/https://deepmind.google/blog/alphago-zero-starting-from-scratch/","archivedAt":"2026-09-24","accessedAt":"2026-09-28","reuse":{"attribution":"David Silver; Demis Hassabis. AlphaGo Zero: Starting from scratch.","licence":"No redistribution licence asserted.","note":"Short attributed paraphrases only; original publication is linked, not redistributed."}},{"id":"source-rlo-2019-deceptive-alignment","title":"Deceptive Alignment (Risks from Learned Optimization, part 4)","authors":["Evan Hubinger","Chris van Merwijk","Vladimir Mikulik","Joar Skalse","Scott Garrabrant"],"publisher":"Alignment Forum and LessWrong","publicationDate":"2019-06-05","version":"Text checked in the post’s first stored version through the LessWrong GraphQL API on 2026-10-06.","canonicalUrl":"https://www.alignmentforum.org/posts/zthDPAjh9w6Ytbeks/deceptive-alignment","evidenceUrl":"https://www.alignmentforum.org/posts/zthDPAjh9w6Ytbeks/deceptive-alignment","language":"en","archivedUrl":"https://web.archive.org/web/20190923160252/https://www.alignmentforum.org/posts/zthDPAjh9w6Ytbeks/deceptive-alignment","archivedAt":"2019-09-23","accessedAt":"2026-10-06","reuse":{"attribution":"Evan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, Scott Garrabrant, \"Deceptive Alignment (Risks from Learned Optimization, part 4)\", Alignment Forum and LessWrong, 2019-06-05.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-extropians-1997-09-26-yudkowsky-4420","title":"Re: Singularity: Generation gap (Extropians mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"Extropians mailing list","publicationDate":"1997-09-26","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 4553a5e84750edd9528beb9e2c297772caee01b17f1653ee34bdae4f8ca62354); 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.3Q97/4420.html","evidenceUrl":"http://extropians.weidai.com/extropians.3Q97/4420.html","language":"en","archivedUrl":"https://web.archive.org/web/20261005144258/https://extropians.weidai.com/extropians.3Q97/4420.html","archivedAt":"2026-10-05","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: Singularity: Generation gap”, Extropians mailing list, 1997-09-26.","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-sl4-2001-02-25-yudkowsky-0102-0594","title":"Re: Military in or out? (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2001-02-25","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 118185c7799c0d30fa1cb5855fe5bdfc4ffbfb8f57184d9006a9e525a33fd204); 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/0102/0594.html","evidenceUrl":"http://sl4.org/archive/0102/0594.html","language":"en","archivedUrl":"https://web.archive.org/web/20260116032126/http://sl4.org/archive/0102/0594.html","archivedAt":"2026-01-16","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: Military in or out?”, SL4 mailing list, 2001-02-25.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-eu-ai-act-2024","title":"Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act)","authors":["European Parliament and Council of the European Union"],"publisher":"Official Journal of the European Union (EUR-Lex)","publicationDate":"2024-07-12","version":"Official Journal L series edition of 12.7.2024, English HTML, retrieved 2026-10-10. The text of the Internet Archive capture of 2024-07-12 matched it apart from the numbering of footnote markers in the amending articles. Later amendments (Regulation (EU) 2026/1744) are not applied to this text.","canonicalUrl":"https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","evidenceUrl":"https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","language":"en","archivedUrl":"https://web.archive.org/web/20240712193938/https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","archivedAt":"2024-07-12","accessedAt":"2026-10-10","reuse":{"attribution":"European Parliament and Council of the European Union, Regulation (EU) 2024/1689 (Artificial Intelligence Act), Official Journal L, 12.7.2024.","licence":"Official legal text; no redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-eloundou-et-al-2023-gpts-are-gpts","title":"GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models","authors":["Tyna Eloundou","Sam Manning","Pamela Mishkin","Daniel Rock"],"publisher":"arXiv:2303.10130v1","publicationDate":"2023-03-17","version":"Full text (v1). Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 f3c91bc909258b4824e2ea101a4f489aaea75d422adaf8508a79883b0d0f74bd. Journal version not retrieved. Eloundou, T., Manning, S., Mishkin, P., & Rock, D. (2024). GPTs are GPTs: Labor market impact potential of LLMs. Science, 384(6702), 1306–1308. https://doi.org/10.1126/science.adj0998","canonicalUrl":"https://arxiv.org/pdf/2303.10130v1","evidenceUrl":"https://arxiv.org/pdf/2303.10130v1","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Tyna Eloundou, Sam Manning, Pamela Mishkin, Daniel Rock, “GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models”, arXiv:2303.10130v1, 2023-03-17.","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-hanson-2008-friendly-projects-vs-products-miri-2013","title":"Friendly Projects vs. Products","authors":["Robin Hanson"],"publisher":"Machine Intelligence Research Institute, The Hanson-Yudkowsky AI-Foom Debate (2013)","publicationDate":"2013","version":"Chapter 42 of The Hanson-Yudkowsky AI-Foom Debate (Machine Intelligence Research Institute, 2013), which compiles the 2008 Overcoming Bias debate with later posts; the compilation heads the post with its date. Current MIRI PDF, retrieved 2026-10-11. SHA-256 6b6dfd0416402e1776d7ddcece95298df2df0ed9bc0bd4a6f4bd81c4b1e5102a.","canonicalUrl":"https://intelligence.org/files/AIFoomDebate.pdf","evidenceUrl":"https://intelligence.org/files/AIFoomDebate.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260929131419/https://intelligence.org/files/AIFoomDebate.pdf","archivedAt":"2026-09-29","accessedAt":"2026-10-11","reuse":{"attribution":"Robin Hanson, “Friendly Projects vs. Products”, in Machine Intelligence Research Institute, The Hanson-Yudkowsky AI-Foom Debate (2013).","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-minimax-prospectus","title":"MiniMax Group Inc. — Global Offering","authors":["MiniMax Group Inc."],"publisher":"MiniMax","publicationDate":"2025-12-31","version":"Dated document, retrieved on 2026-10-15 (SHA-256 7853021e46da5714242f8d27984498cbfef3e260d5a7cbb8d4e6d62d83ab9df3) directly from the publisher. Publisher as printed on the inspected expression: MiniMax Group Inc. (via HKEXnews).","canonicalUrl":"https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1231/2025123100025.pdf","evidenceUrl":"https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1231/2025123100025.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260930092156/https://www1.hkexnews.hk/listedco/listconews/sehk/2025/1231/2025123100025.pdf","archivedAt":"2026-09-30","accessedAt":"2026-10-15","reuse":{"attribution":"MiniMax Group Inc., “MiniMax Group Inc. — Global Offering”, 2025-12-31.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-gs-tc-achiam-2026","title":"OpenAI disbands mission alignment team","authors":["Lucas Ropek"],"publisher":"TechCrunch","publicationDate":"2026-02-11","version":"Dated page, retrieved on 2026-10-15 (SHA-256 4b410eaf838ea986267156a86a8068e632dc2691e235546aa45e7b7649af8628) directly from the publisher.","canonicalUrl":"https://techcrunch.com/2026/02/11/openai-disbands-mission-alignment-team-which-focused-on-safe-and-trustworthy-ai-development/","evidenceUrl":"https://techcrunch.com/2026/02/11/openai-disbands-mission-alignment-team-which-focused-on-safe-and-trustworthy-ai-development/","language":"en","archivedUrl":"https://web.archive.org/web/20260827200053/https://techcrunch.com/2026/02/11/openai-disbands-mission-alignment-team-which-focused-on-safe-and-trustworthy-ai-development/","archivedAt":"2026-08-27","accessedAt":"2026-10-15","reuse":{"attribution":"Lucas Ropek, “OpenAI disbands mission alignment team”, 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":"c2-src-minimax-m1-abs-v1","title":"MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention (arXiv abstract page, v1)","authors":["MiniMax"],"publisher":"arXiv","publicationDate":"2025-06-16","version":"Dated document, retrieved on 2026-10-27 (SHA-256 02651f9afdf37a63103be751b903c962a65c0cda8f57399f18bc6719726cc64f) directly from the publisher. Version: 2506.13585v1. Abstract page of version 1 (submitted 16 June 2025); later versions not inspected.","canonicalUrl":"https://arxiv.org/abs/2506.13585v1","evidenceUrl":"https://arxiv.org/abs/2506.13585v1","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"MiniMax, “MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention (arXiv abstract page, v1)”, 2025-06-16.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-bletchley-chair-safety-testing-2023","title":"Safety Testing: Chair's Statement of Session Outcomes, 2 November 2023","authors":["Chair of the AI Safety Summit (UK government)"],"publisher":"Department for Science, Innovation and Technology","publicationDate":"2023-11-02","version":"Retrieved on 2026-10-27 (SHA-256 c7b3a9cb614c69a77061b04735d88697021a21ad583ba738a5b11befe7d93e38). GOV.UK HTML policy paper, 'Published 2 November 2023', as served on 2026-10-27. Publisher as printed on the inspected expression: GOV.UK (Prime Minister's Office, Foreign, Commonwealth & Development Office, Department for Science, Innovation and Technology).","canonicalUrl":"https://www.gov.uk/government/publications/ai-safety-summit-2023-chairs-statement-safety-testing-2-november/safety-testing-chairs-statement-of-session-outcomes-2-november-2023","evidenceUrl":"https://www.gov.uk/government/publications/ai-safety-summit-2023-chairs-statement-safety-testing-2-november/safety-testing-chairs-statement-of-session-outcomes-2-november-2023","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"Chair of the AI Safety Summit (UK government), “Safety Testing: Chair's Statement of Session Outcomes, 2 November 2023”, 2023-11-02.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-uk-google-deepmind-mou-2025","title":"Memorandum of Understanding between the UK and Google DeepMind on AI opportunities and security","authors":["Department for Science, Innovation and Technology"],"publisher":"Department for Science, Innovation and Technology","publicationDate":"2025-12-11","version":"Retrieved on 2026-10-27 (SHA-256 e89909172b1ba90867e13c78c8d34c46890afbf362a061fe80c0b440a119fe6f). GOV.UK HTML text, 'Published 11 December 2025'; signatory names given without a signing date. 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-google-deepmind-on-ai-opportunities-and-security/memorandum-of-understanding-between-the-uk-and-google-deepmind-on-ai-opportunities-and-security","evidenceUrl":"https://www.gov.uk/government/publications/memorandum-of-understanding-between-the-uk-and-google-deepmind-on-ai-opportunities-and-security/memorandum-of-understanding-between-the-uk-and-google-deepmind-on-ai-opportunities-and-security","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"Department for Science, Innovation and Technology, “Memorandum of Understanding between the UK and Google DeepMind on AI opportunities and security”, 2025-12-11.","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-uk-google-deepmind-mou-page","title":"Memorandum of Understanding between the UK and Google DeepMind on AI opportunities and security (publication page)","authors":["Department for Science, Innovation and Technology"],"publisher":"Department for Science, Innovation and Technology","publicationDate":"2025-12-11","version":"Retrieved on 2026-10-27 (SHA-256 b17524e7766e925290385cfe1fc4392209bef879724f0ab4aa9570548d828fcd). Publication page as served on 2026-10-27 (Published 11 December 2025) Its wording can change; earlier wording is not authenticated. Publisher as printed on the inspected expression: GOV.UK.","canonicalUrl":"https://www.gov.uk/government/publications/memorandum-of-understanding-between-the-uk-and-google-deepmind-on-ai-opportunities-and-security","evidenceUrl":"https://www.gov.uk/government/publications/memorandum-of-understanding-between-the-uk-and-google-deepmind-on-ai-opportunities-and-security","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"Department for Science, Innovation and Technology, “Memorandum of Understanding between the UK and Google DeepMind on AI opportunities and security (publication page)”, 2025-12-11.","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-us-eo-14110-fr","title":"Executive Order 14110 of October 30, 2023: Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence (88 FR 75191)","authors":["The White House"],"publisher":"The White House","publicationDate":"2023-11-01","version":"Retrieved on 2026-10-27 (SHA-256 46739dfc1af256961458d3b73fa4d6f4b447b38eaa72efd80966463818be68ab). Federal Register print of 1 November 2023, Vol. 88, No. 210, pp. 75191–75226, FR Doc. 2023-24283. The PDF text layer renders superscript exponents inline (10^26 appears as \"1026\"); quotes stop before such numbers. 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-2023-11-01/pdf/2023-24283.pdf","evidenceUrl":"https://www.govinfo.gov/content/pkg/FR-2023-11-01/pdf/2023-24283.pdf","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"The White House, “Executive Order 14110 of October 30, 2023: Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence (88 FR 75191)”, 2023-11-01.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cpo-ch-bis-uae-2026","title":"Enhanced Favorable Treatment for the United Arab Emirates Under the Export Administration Regulations (91 FR 43034)","authors":["Bureau of Industry and Security, U.S. Department of Commerce"],"publisher":"Bureau of Industry and Security, in the Federal Register (govinfo.gov)","publicationDate":"2026-07-14","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 4e4b5555a41f7799f4e4dfe192a3a159c6fbb023da36ffacd76e4bbb80c5d9d5). Federal Register print of Tuesday, July 14, 2026, Vol. 91, No. 133, beginning at p. 43034, FR Doc. 2026-14132 (filed 7-10-26). Final rule. Retrieved again at admission on 2026-10-30 (SHA-256 4e4b5555a41f7799f4e4dfe192a3a159c6fbb023da36ffacd76e4bbb80c5d9d5, the same bytes).","canonicalUrl":"https://www.govinfo.gov/content/pkg/FR-2026-07-14/pdf/2026-14132.pdf","evidenceUrl":"https://www.govinfo.gov/content/pkg/FR-2026-07-14/pdf/2026-14132.pdf","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Bureau of Industry and Security, U.S. Department of Commerce, “Enhanced Favorable Treatment for the United Arab Emirates Under the Export Administration Regulations (91 FR 43034)”, 2026-07-14.","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-cpo-dc-google-texas-40bn-2025","title":"Google’s investments run deep in the heart of Texas (Google announces new $40 billion investment in Texas)","authors":["Google (The Keyword blog)"],"publisher":"Google (The Keyword blog)","publicationDate":"2025-11-14","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 a94786832acde10b267b5f7b1e6908636f857952aef44b47b0c0f7244cb35ac9). Blog post as served on 2026-10-28; dated \"Nov 14, 2025\". Retrieved again at admission on 2026-10-30 (SHA-256 01b171995581cc791cb6ee2b7554e310d31164f292af8ff9db1ecab06fbce325). Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://blog.google/company-news/inside-google/company-announcements/google-american-innovation-texas/","evidenceUrl":"https://blog.google/company-news/inside-google/company-announcements/google-american-innovation-texas/","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Google (The Keyword blog), “Google’s investments run deep in the heart of Texas (Google announces new $40 billion investment in Texas)”, 2025-11-14.","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-cpo-pw-actionnews5-xai-permit-2025","title":"Health dept. grants permit for xAI turbines","authors":["Action News 5 (WMC, Memphis)"],"publisher":"Action News 5 (WMC, Memphis)","publicationDate":"2025-07-02","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 761f20210c26fff85d89c72a88ee7b4b807e26d68c671e88144c3b2b82972377). Article as served on 2026-10-28; 'Published: Jul. 2, 2025 at 5:01 PM CDT | Updated: Jul. 2, 2025 at 10:26 PM CDT'. The text says the approval came 'on Thursday', but 2 July 2025 was a Wednesday; the date is taken from the publication line. The Shelby County Health Department's permit was not retrieved. Retrieved again at admission on 2026-10-30 (SHA-256 89c69d9c24663ef210786cb058d978d41985ba3b38c591c69428d2c7752d06d9).","canonicalUrl":"https://www.actionnews5.com/2025/07/02/health-dept-grants-permit-xai-turbines/","evidenceUrl":"https://www.actionnews5.com/2025/07/02/health-dept-grants-permit-xai-turbines/","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Action News 5 (WMC, Memphis), “Health dept. grants permit for xAI turbines”, 2025-07-02.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cpo-shelby-permit-01156-01pc","title":"Construction Permit 01156-01PC (scanned)","authors":["Shelby County Health Department"],"publisher":"Shelby County Health Department","publicationDate":"2025-07-02","version":"Not read at admission: the county’s address answered HTTP 404 and the Internet Archive refused this admission. The cross-family check (2026-10-29) read the capture linked here and inspected page 1 of the scan visually; the record’s details rest on that reading.","canonicalUrl":"https://www.shelbytnhealth.com/DocumentCenter/View/7174/01156-01PC-070225Permit","evidenceUrl":"https://www.shelbytnhealth.com/DocumentCenter/View/7174/01156-01PC-070225Permit","language":"en","archivedUrl":"https://web.archive.org/web/20250703012450/https://www.shelbytnhealth.com/DocumentCenter/View/7174/01156-01PC-070225Permit","archivedAt":"2025-07-03","accessedAt":"2026-10-30","reuse":{"attribution":"Shelby County Health Department, “Construction Permit 01156-01PC (scanned)”, 2025-07-02.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-ab-gov-bwc-msp2025-wp28","title":"Statement by Japan (BWC/MSP/2025/WP.28)","authors":["United Nations (document of the 2025 Meeting of States Parties to the Biological Weapons Convention), submitted by Japan"],"publisher":"United Nations (document of the 2025 Meeting of States Parties to the Biological Weapons Convention), submitted by Japan","publicationDate":null,"version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 8b662b8ff4896d9509c4c5cefbe7bc40b551573fc5f032ae85fa349c1d4e2ed6). English PDF of document BWC/MSP/2025/WP.28, dated 17 December 2025; 2025 Meeting, Geneva, 15-17 December 2025. Retrieved 2026-10-28 through the documents.un.org symbol access endpoint. Retrieved again at admission on 2026-10-30 (SHA-256 8b662b8ff4896d9509c4c5cefbe7bc40b551573fc5f032ae85fa349c1d4e2ed6, the same bytes).","canonicalUrl":"https://documents.un.org/doc/undoc/gen/g25/211/13/pdf/g2521113.pdf","evidenceUrl":"https://documents.un.org/doc/undoc/gen/g25/211/13/pdf/g2521113.pdf","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"United Nations (document of the 2025 Meeting of States Parties to the Biological Weapons Convention), submitted by Japan, “Statement by Japan (BWC/MSP/2025/WP.28)”.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}}],"evidence":[{"id":"evidence-lineage-ii-concrete-discusses-oversight","sourceRole":"primary","support":"supports","locator":"PDF §5, pp. 11–12","passage":"The section considers learning with expensive detailed evaluation and cheaper imperfect feedback.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-25","note":"Direct source inspection; machine check only. Discussion of one formulation; not earliest terminology or equivalence with later superhuman-evaluation proposals."},{"id":"evidence-eval-i-metr-o1-report","sourceRole":"primary","support":"supports","locator":"Displayed date; opening paragraph; Details of Evaluation Procedure > Model Access.","passage":"METR dates o1-preview access to September 3 and its evaluation end to September 9.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-27","note":"Original publisher passage machine-inspected. This is not human review or independent reproduction of the reported activity."},{"id":"evidence-atlas-alphago-rl","sourceRole":"primary","support":"supports","locator":"Paragraph beginning “It is able to do this” and following paragraph","passage":"The account combines a neural network, search and repeated self-play updates.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-28","note":"Original source passage checked by the machine agent that proposed the record. No independent machine audit or human source inspection is recorded for this increment."},{"id":"evidence-prov-rsp-v3-3-marginal-risk-board-ltbt-approval","sourceRole":"primary","support":"supports","locator":"PDF p. 13 (printed p. 12), §3.4 Procedures, item 5 ('Modified process when marginal risk analysis is important to our case')","passage":"In the event that marginal risk analysis (see previous section) plays a major role in a decision to move forward, explicit approval of the Risk Report by the Board and LTBT (rather than just the CEO and RSO) will be required.","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 2026-05-28. Extraction note: Wording identical to v3.1 (and to v3.2, not proposed there: quotation limit). Human editorial review is pending."},{"id":"evidence-prov-rsp-v3-3-automated-rd-threshold","sourceRole":"primary","support":"supports","locator":"PDF p. 9 (printed p. 9), §1 Our Recommendations for Industry-Wide Safety, table row 'Automated R&D in key domains', left column ('Capability or usage threshold'), paragraph beginning 'We would consider scenario (2)', to the end of condition (a)","passage":"We would consider scenario (2) to have occurred where (a) we observe or expect double the rate of progress8 in AI aggregate capabilities compared to the rate we’d expect in the absence of significant AI contributions to AI R&D","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 2026-05-28. Extraction note: The plain text layer interleaves the table's three columns line by line, so the quote was verified against the left column extracted on its own (pdftotext with a right margin of 572 points). That layer has no word space between “expect” and “double”; the quote restores it. “8” is the superscript footnote marker. Quote ends before condition (b). Human editorial review is pending."},{"id":"evidence-provchange-preparedness-v2-final-decision-authority-later","sourceRole":"primary","support":"supports","locator":"PDF p. 4 (printed p. 3), §1 Introduction, final paragraph, second sentence","passage":"Preparedness Framework Version 2 states that OpenAI Leadership can approve or reject the Safety Advisory Group’s recommendations and that the Board’s Safety and Security Committee provides oversight of these decisions.","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-preparedness-v2-final-decision-authority-earlier","sourceRole":"primary","support":"supports","locator":"PDF p. 22 (printed p. 22), Governance › “Operations”, opening paragraph, second sentence, items (2) and (3)","passage":"Preparedness Framework (Beta) states that there is an advisory group, the Safety Advisory Group, to provide nuanced input and recommendations, and a final decision-maker, OpenAI Leadership, with the option for the OpenAI Board of Directors to overrule.","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-yk-t01-e2","sourceRole":"primary","support":"supports","locator":"message body, paragraph beginning 'I think of the Singularity'","passage":"would not suffice to keep *me* in … Pull the plug? Hey, I'm already running on 200,000 computers","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-07","note":"Candidate pack passage T01-e2. The quotation (18 words; trimmed 2026-10-13 from 26 words) matched the retrieved expression (SHA-256 4553a5e84750edd9528beb9e2c297772caee01b17f1653ee34bdae4f8ca62354) when checked on 2026-10-07."},{"id":"evidence-yk-t15-s1","sourceRole":"primary","support":"supports","locator":"message body, paragraph beginning 'Q: \"Can non-Friendly AI or toollike AI be built?\"'","passage":"Q: \"Can non-Friendly AI or toollike AI be built?\" A: \"Not safely.\"","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-07","note":"Candidate pack passage T15-s1. The quotation (12 words) matched the retrieved expression (SHA-256 118185c7799c0d30fa1cb5855fe5bdfc4ffbfb8f57184d9006a9e525a33fd204) when checked on 2026-10-07."},{"id":"evidence-ec-m03-10","sourceRole":"primary","support":"supports","locator":"pdf page 1; Abstract","passage":"We conclude that Generative Pre-trained Transformers exhibit characteristics of general-purpose technologies (GPTs), suggesting that as these models could have notable economic, social, and policy implications.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M03-10 (research/economics-mainstream). The quotation (25 words) matched the retrieved current copy (SHA-256 f3c91bc909258b4824e2ea101a4f489aaea75d422adaf8508a79883b0d0f74bd) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m03-11","sourceRole":"primary","support":"supports","locator":"pdf page 11; 4 Results, 4.1 Summary Statistics (printed p. 10 begins section 4.1 on preceding page)","passage":"Although the potential for tasks to be affected is extensive, GPTs must be incorporated into broader systems to realize this potential fully. As is common with general-purpose technologies, such co-invention barriers may impede the rapid diffusion of GPTs into economic applications.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M03-11 (research/economics-mainstream). The quotation (41 words) matched the retrieved current copy (SHA-256 f3c91bc909258b4824e2ea101a4f489aaea75d422adaf8508a79883b0d0f74bd) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"lg-ev-gs-mission-alignment-team-disbanded-2026-a","sourceRole":"secondary","support":"supports","locator":"Paragraph 1","passage":"OpenAI has disbanded a team that was designed to communicate the company’s mission to the public and to its own employees.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (21 words) matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-gs-mission-alignment-team-disbanded-2026","sourceRole":"secondary","support":"supports","locator":"Paragraph 2","passage":"OpenAI confirmed to TechCrunch that the team’s members have now been assigned to other roles.","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 directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ap-scott-garrabrant-rlo-2019-deceptive-alignment","sourceRole":"primary","support":"supports","locator":"Post author list in the publisher’s post data (LessWrong GraphQL API)","passage":"Scott Garrabrant","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval through the publisher’s post data and checked by a reviewer from a different model family on 2026-10-27. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"evidence-ap-robin-hanson-hanson-2008-friendly-projects-vs-products","sourceRole":"primary","support":"supports","locator":"PDF text layer, first two pages: author name on the title page or first page","passage":"Robin Hanson","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"evidence-ap-pamela-mishkin-eloundou-et-al-2023-gpts-are-gpts","sourceRole":"primary","support":"supports","locator":"PDF text layer, first two pages: author name on the title page or first page","passage":"Pamela Mishkin","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval and checked by a reviewer from a different model family on 2026-10-27. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"ev-cmp-rsp-v2-1-d1-cyber-1","sourceRole":"primary","support":"supports","locator":"PDF p. 9 (printed p. 5), §2, table 'Capabilities / Ongoing Assessment', row 'Cyber Operations'","passage":"The edition lists cyber operations under “Ongoing Assessment”: the capacity to substantially strengthen or automate sophisticated, destructive cyberattacks.","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-faif-2026-d2-definition-1","sourceRole":"primary","support":"supports","locator":"PDF p. 5, first §2.3 'Systemic risk acceptance determination', first sentence, up to 'acceptable'","passage":"The edition says xAI applies systemic risk acceptance criteria, with a margin of security, to every identified risk, to decide whether each such risk, and systemic risk overall, is acceptable.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Explicit paraphrase, written on 2026-10-27 by an automated research assistant from the comparison’s quotation of this passage, which was checked on 2026-10-15 against the publisher copy (SHA-256 recorded in the comparison pack). The edition could not be retrieved again on 2026-10-27; the archive capture of 2026-07-16 has the same SHA-256 (research/archive-checks-2026-10-26/codex-report.md). Given as a paraphrase to keep the edition within the quotation limit. Human editorial review is pending."},{"id":"c2-ev-minimax-m1-report","sourceRole":"primary","support":"supports","locator":"arXiv abstract page, submission line","passage":"Submitted on 16 Jun 2025","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (5 words), a part of a longer passage that the pack joined across separate fields of the page, matched the retrieved text on 2026-10-27 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 pack had joined this field with others of the same page into one quotation; at admission each field became its own quotation (cross-family check, 2026-10-27). No human editorial review.","retrievalRoute":"direct"},{"id":"c2-ev-minimax-m1-report-2","sourceRole":"primary","support":"supports","locator":"Abstract","passage":"We publicly release MiniMax-M1 at this https URL.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (8 words) matched the retrieved text on 2026-10-27 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":"c2-ev-minimax-m1-report-pros","sourceRole":"primary","support":"context","locator":"Business, “Large Language Model: MiniMax M Series”","passage":"MiniMax-M1, launched in June 2025, is an open-source, large-scale hybrid-attention reasoning model.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (12 words) matched the retrieved text on 2026-10-27 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":"evidence-fl-intl-bletchley-chair-testing-published-1","sourceRole":"primary","support":"supports","locator":"Document header","passage":"Published 2 November 2023","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (4 words) matched the text retrieved for the candidate pack on 2026-10-27 and again at admission on 2026-10-28, in a fresh retrieval of the same address (the page’s bytes had changed). Read directly from www.gov.uk. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. The header names the three publishing bodies. A cross-family machine check (2026-10-28, item 6) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-uk-gdm-mou-published-1","sourceRole":"primary","support":"supports","locator":"Publication page metadata and details","passage":"Paraphrase: 'Published: 11 December 2025'; the MoU 'sets out the agreement signed by Liz Kendall MP, Secretary of State for Science, Innovation and Technology and Sir Demis Hassabis, CEO of Google DeepMind'.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The paraphrase is Selekoa’s, written from the text retrieved on 2026-10-27 and read directly from www.gov.uk; it is not a quotation and was not machine-matched. A cross-family machine check (2026-10-28, item 68) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-uk-gdm-mou-published-2","sourceRole":"primary","support":"supports","locator":"General terms","passage":"Paraphrase: the MoU is between the Secretary of State for DSIT, acting on behalf of the Crown, and Google LLC (Google DeepMind).","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The paraphrase is Selekoa’s, written from the text retrieved on 2026-10-27 and read directly from www.gov.uk; it is not a quotation and was not machine-matched. A cross-family machine check (2026-10-28, item 68) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-ch-bis-uae-2026-published-1","sourceRole":"primary","support":"supports","locator":"91 FR 43034, masthead","passage":"Federal Register / Vol. 91, No. 133 / Tuesday, July 14, 2026 / Rules and Regulations","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (13 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-ch-bis-uae-2026-published-2","sourceRole":"primary","support":"supports","locator":"91 FR 43034, DATES","passage":"This rule is effective July 10, 2026.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). 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. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-dc-google-texas-40bn-1","sourceRole":"primary","support":"supports","locator":"paragraph 1","passage":"a new $40 billion investment in the Lone Star State through 2027","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (12 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval (SHA-256 01b171995581cc791cb6ee2b7554e310d31164f292af8ff9db1ecab06fbce325; the bytes differ from the pack’s retrieval). Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-dc-google-texas-40bn-2","sourceRole":"primary","support":"supports","locator":"paragraph 1","passage":"the funding will help build new cloud and AI infrastructure, including new data center campuses in Armstrong and Haskell Counties.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The paraphrase is Selekoa’s, written from the text retrieved for the candidate pack on 2026-10-28; it is not a quotation and was not machine-matched. The cross-family check found it faithful. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-shelby-xai-turbine-permit-1","sourceRole":"secondary","support":"context","locator":"paragraph beginning “MEMPHIS, Tenn. (WMC) - The Shelby…”","passage":"The Shelby County Health Department approved xAI’s construction of natural gas-fired combustion turbines","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (13 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval (SHA-256 89c69d9c24663ef210786cb058d978d41985ba3b38c591c69428d2c7752d06d9; the bytes differ from the pack’s retrieval). Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 58) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-shelby-xai-turbine-permit-2","sourceRole":"secondary","support":"context","locator":"paragraph beginning “In the permit, 15 turbines were…”","passage":"In the permit, 15 turbines were approved to be permanently built.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (11 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval (SHA-256 89c69d9c24663ef210786cb058d978d41985ba3b38c591c69428d2c7752d06d9; the bytes differ from the pack’s retrieval). Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 58) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-shelby-xai-turbine-permit-3","sourceRole":"secondary","support":"supports","locator":"paragraph beginning “The health department also issued a…”","passage":"temporary turbine engines, which are scheduled to be removed by the end of July","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (14 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval (SHA-256 89c69d9c24663ef210786cb058d978d41985ba3b38c591c69428d2c7752d06d9; the bytes differ from the pack’s retrieval). Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 58) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-shelby-xai-turbine-permit-4","sourceRole":"secondary","support":"context","locator":"paragraph beginning “SELC, alongside the NAACP…”","passage":"the report says SELC, alongside the NAACP, intend to sue xAI over its use of the gas turbines.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The paraphrase is Selekoa’s, written from the text retrieved for the candidate pack on 2026-10-28; it is not a quotation and was not machine-matched. The cross-family check found it faithful. A cross-family machine check (2026-10-29/30, item 58) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-pw-shelby-xai-turbine-permit-5","sourceRole":"primary","support":"supports","locator":"Construction Permit 01156-01PC, PDF p. 1","passage":"The permit, issued on 2 July 2025 to CTC Property LLC, covers fifteen Solar SMT-130 natural gas-fired combustion turbines of 16.48 MW each and prints an expiry of 2 January 2027.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-30","note":"Added at the admission of the compute, power and orbit pack (research/compute-power-orbit). The regulator’s own permit, as the cross-family check read it in an Internet Archive capture of 3 July 2025 (visual inspection of the scan); not read at admission. The paraphrase is Selekoa’s; it is not a quotation and was not machine-matched. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-gov-us-eo14110-sec44aii-academies-study-1","sourceRole":"primary","support":"supports","locator":"Sec. 4.4(a)(ii)(A), 88 FR 75201","passage":"assesses the ways in which AI can increase biosecurity risks, including risks from generative AI models trained on biological data, and makes recommendations on how to mitigate these risks","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (29 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 subsection opens by directing the Secretary of Defense, within 120 days, to contract with the National Academies of Sciences, Engineering, and Medicine for the study. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-gov-eu-ai-act-annex13-biological-sequences-1","sourceRole":"primary","support":"supports","locator":"Annex XIII, point (d)","passage":"and the specific type of inputs and outputs (e.g. biological sequences)","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (11 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. The closing clause of point (d), whose earlier clauses cover the input and output modalities of the model and the state of the art thresholds for determining high-impact capabilities for each modality. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-gov-eu-ai-act-annex13-biological-sequences-2","sourceRole":"primary","support":"context","locator":"Annex XIII, heading and opening words","passage":"Criteria for the designation of general-purpose AI models with systemic risk referred to in Article 51 … the Commission shall take into account the following criteria","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (25 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 Annex's own heading and the end of its opening sentence, which begins 'For the purpose of determining that a general-purpose AI model has capabilities or an impact equivalent to those set out in Article 51(1), point (a)'. Cited for the framing of the criteria, which the statement now follows. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-gov-bwc-japan-wp28-misuse-risk-1","sourceRole":"primary","support":"supports","locator":"Paragraph 5","passage":"New technologies such as genome editing technology, synthetic biology, and artificial intelligence offer great benefits to humanity; however, if misused, they could lead to the production of biological weapons that cause devastating harm.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (33 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 paragraph continues that there are growing concerns that these technologies are becoming available to a wider range of actors, increasing the risk of proliferation of biological weapons. No human editorial review.","retrievalRoute":"direct"}],"at":{"entities":[0,39,43,48,53,54,57,104,107,109,115,126,132,233,248,331,350,364,394,459,491,515,564,609,757,882,884,902,979,1019,1034,1038,1064,1090,1165,1204],"claims":[235,266,297,426,521,599,993,1088,1347,1848,2238,2261,2294,2506,2687,2787,2943,3039,3073,3132,3200,3328,3330,3349],"sources":[37,42,48,49,52,103,116,132,336,361,426,457,517,558,582,666,805,850,869,873,874,914,979,1021,1033,1043],"evidence":[256,287,318,458,587,677,678,1219,1324,1679,1680,2379,2380,2885,2908,2941,3153,3374,3537,3538,3539,3802,3946,3947,4034,4035,4144,4145,4340,4341,4342,4343,4344,4573,4576,4577,4598]},"version":"1lg47t7"}