{"entities":[{"id":"anthropic","name":"Anthropic"},{"id":"google-brain","name":"Google Brain"},{"id":"greg-yang","name":"Greg Yang"},{"id":"rsp-v2-2","name":"Anthropic Responsible Scaling Policy — v2.2"},{"id":"rsp-v3-2","name":"Anthropic Responsible Scaling Policy — v3.2"},{"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":"frontier-governance-2026","name":"OpenAI Frontier Governance Framework — May 2026 edition"},{"id":"faif-2026","name":"xAI Frontier Artificial Intelligence Framework — June 2026 edition"},{"id":"durk-kingma","name":"Durk Kingma"},{"id":"tensor-programs-v","name":"Tensor Programs V (2022 paper)"},{"id":"ms-fgf-2026","name":"Microsoft Frontier Governance Framework — February 2026"},{"id":"magic-ai","name":"Magic"},{"id":"extropians-1998-06-23-den-otter","name":"The AI revolution (Was: Re: >H ART: The Truman Show)"},{"id":"sl4-2000-07-07-atkins","name":"AI testing and containment Re: [SL4] Programmed morality"},{"id":"ai-positive-negative-factor-global-risk","name":"Artificial Intelligence as a Positive and Negative Factor in Global Risk"},{"id":"race-compression","name":"Strategic or race compression"},{"id":"observed-behaviour-as-evidence","name":"Observed behaviour under training or oversight as evidence"},{"id":"switch-off-if-dangerous","name":"Switching it off if it becomes dangerous"},{"id":"ai-solves-alignment","name":"AI systems will solve alignment for us"},{"id":"agi-decades-away","name":"AGI is decades away, so the problem can wait"},{"id":"no-body-no-danger","name":"Without a body it cannot do much"},{"id":"extropians-1996-12-08-yudkowsky-3761","name":"Re: Trans- vs. Non-human"},{"id":"lw-2017-theres-no-fire-alarm-for-artificial","name":"There's No Fire Alarm for Artificial General Intelligence"},{"id":"lw-2021-discussion-with-eliezer-yudkowsky-on-agi","name":"Discussion with Eliezer Yudkowsky on AGI interventions"},{"id":"cn-generative-ai-interim-measures-2023","name":"Interim Measures for the Management of Generative Artificial Intelligence Services (生成式人工智能服务管理暂行办法), Order No. 15"},{"id":"unautomated-tasks-bottleneck-growth","name":"Essential unautomated tasks bottleneck growth"},{"id":"ai-as-task-automation","name":"AI as the automation of tasks"},{"id":"machines-cheaper-at-every-task","name":"Machines cheaper than labour at every task"},{"id":"hanson-1998-some-skepticism","name":"Some Skepticism (comment in \"A Critical Discussion of Vinge's Singularity Concept\")"},{"id":"clancy-besiroglu-2023-great-inflection","name":"The Great Inflection? A Debate About AI and Explosive Growth"},{"id":"erdil-besiroglu-2023-explosive-growth-review","name":"Explosive growth from AI automation: A review of the arguments"},{"id":"calvano-et-al-2020-algorithmic-pricing-collusion","name":"Artificial Intelligence, Algorithmic Pricing, and Collusion"},{"id":"nourse-1955-automation-hearings-statement","name":"Statement (former Chairman, Council of Economic Advisers), in Automation and Technological Change: Hearings"},{"id":"ai-improving-ai-accelerates-capability","name":"Machine intelligence improving itself accelerates its capability growth"},{"id":"eric-steinberger","name":"Eric Steinberger"},{"id":"rahul-patil","name":"Rahul Patil"},{"id":"matt-clancy","name":"Matt Clancy"},{"id":"eu-commission-opinion-code-2025","name":"Commission Opinion C(2025) 5361 final on the assessment of the General-Purpose AI Code of Practice"},{"id":"lumen-orbit-wp-2024","name":"Why we should train AI in space"},{"id":"aili-2025-carbon-neutral-data-centres-in-space","name":"The development of carbon-neutral data centres in space"},{"id":"oc3-terrestrial-constraints","name":"Terrestrial constraints as a reason for orbital compute"},{"id":"oc11-sky-and-debris","name":"Harm to the sky and debris risk from orbital data centres"},{"id":"bwc-msp-2025-wp4","name":"Scientific and Technological Advances in the Life Sciences since 1975: How the Last 50 Years Will Shape the BTWC S&T Advisory Mechanism (BWC/MSP/2025/WP.4)"}],"claims":[{"id":"supersedes-fsf-3-1","statement":"Frontier Safety Framework v3.1 replaces v3.0 in the publisher’s documented framework version sequence.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"This records replacement in the public policy version sequence, supported by the publisher’s update history. It does not retire historical claims or prove implementation.","evidenceIds":["evidence-supersedes-fsf-3-1","evidence-supersedes-fsf-3-1-2"]},{"id":"kingma-brain-research-prior","statement":"Durk Kingma’s biography includes Google Brain in a combined Google Brain / DeepMind Research Scientist entry.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Do not assign all six years to Google Brain or equate predecessor DeepMind with Google DeepMind. Institutional boundary dates within this entry remain unresolved.","evidenceIds":["evidence-kingma-brain-research-prior"],"attributes":{"role":"Research Scientist"}},{"id":"xm-yang-tensor-programs","statement":"Greg Yang co-authored Tensor Programs V on hyperparameter transfer across model sizes.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Version-one authorship predates xAI. Neither employment overlap nor co-authorship establishes institutional adoption or transmission.","evidenceIds":["evidence-xm-yang-tensor-programs"]},{"id":"prov-rsp-v2-2-misuse-uplift-threshold","statement":"RSP v2.2 states that its CBRN-3 Capability Threshold is the ability to significantly help individuals or groups with basic technical backgrounds (e.g., undergraduate STEM degrees) create/obtain and deploy CBRN weapons.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the edition states, quoted from the cited expression. It is not evidence of implementation, compliance or effectiveness.","evidenceIds":["evidence-prov-rsp-v2-2-misuse-uplift-threshold"],"attributes":{"provision":{"kind":"threshold","key":"rsp-misuse-uplift-threshold"}}},{"id":"prov-frontier-governance-2026-update-legal-oversight","statement":"Frontier Governance Framework (May 2026) states that OpenAI’s Legal function, in collaboration with relevant internal stakeholders, will oversee the process for FGF updates, including determining which updates are 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-frontier-governance-2026-update-legal-oversight"],"attributes":{"provision":{"kind":"decision_authority","key":"fgf-update-legal-oversight","parties":[{"role":"decides","actor":"OpenAI Legal function (which updates are required)"}]}}},{"id":"ts-a6-denotter-1998","statement":"den Otter’s message of 1998-06-23 argues that in the development of AI, as in any economic or arms race, people will sacrifice safety to get an extra edge.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Offered as a reason why human extinction is a rational worry; it names no developer and no first-mover advantage, and the reply concerns a more productive but less stable AI, so speed is not specified.","evidenceIds":["evidence-ts-a6-denotter-1998"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","qualifier":"Safety traded for competitive advantage."}}},{"id":"ts-b5-atkins-2000","statement":"Atkins’s message of 2000-07-07 says that to really test what an AI will do once it is \"loose\", testers would have to give it a very realistic simulation of the world and watch what it does to the humans.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"About the limits of pre-release testing, without deception; Atkins doubts such a simulation is feasible.","evidenceIds":["evidence-ts-b5-atkins-2000"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction"}}},{"id":"yk-y01-extropians-1996-12-08-yudkowsky-3761-c","statement":"The Extropians message “Re: Trans- vs. Non-human” (1996-12-08) contests Y01. Replying to Steve Witham, who asked why certain values should not be 'inerasably etched' on a newborn Power, he argues that a Power would work against limits imposed on it, turning the constraint into a contest inside the Power's own mind that the Power would almost certainly win.","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 proposal discussed is an unerasable protective goal (he compares it to the Three Laws of Robotics), not an off-switch. The argument rests on his 1996 view that an unconstrained Power would be ethical (see his Extropians messages of 1996-11-22); his conclusion at the time was to leave Powers unconstrained. Quotation shortened to respect the ~10% cap for a short message. The \"leashes\" are an unerasable protect-humans goal, not a shutdown switch; the 1996 argument rests on his then view that a Power should find its own ethics.","evidenceIds":["evidence-yk-t01-e1"],"attributes":{"thought":{"roles":["contesting","earliest-located"],"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-y02-lw-2008-that-tiny-note-of-discord-c","statement":"The post “That Tiny Note of Discord” (2008-09-23) contests Y02. Recapping his 2008 autobiographical series, he states that his 1997 self believed any superintelligence would automatically do what is right, and names the flaw in that belief.","evidenceIds":["evidence-yk-t02-e"]},{"id":"yk-y02-sl4-2000-11-13-yudkowsky-0011-0181-c","statement":"The SL4 message “Re: Ben's \"Extropian Creed\"” (2000-11-13) contests Y02. Replying to Ben Goertzel about his newspaper article on Extropians and poverty, he describes his own work on Friendly AI and contrasts it with his former belief that an objective morality would make an AI friendly without design work.","evidenceIds":["evidence-yk-t02-s1"]},{"id":"yk-y02-extropians-1999-08-01-yudkowsky-1386-c","statement":"The Extropians message “Re: IA vs. AI was: longevity vs singularity” (1999-08-01) contests Y02. In a long reply to den Otter, he points to his old Extropians posts and the old version of 'Staring Into the Singularity' as showing how he moved from assuming superintelligences would be good and nice to humans to not knowing what absolute good is.","evidenceIds":["evidence-yk-t02-e3"]},{"id":"yk-y07-lw-2021-discussion-with-eliezer-yudkowsky-on-agi-c","statement":"The post “Discussion with Eliezer Yudkowsky on AGI interventions” (2021-11-11) contests Y07. Answering the suggestion that humans could at least check AI-generated arguments for a design, he argues that an untrustworthy superintelligence could fool or covertly manipulate the checkers, so checking presupposes a belief in its alignment obtained some other way.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: We can verify its alignment output. The quotation is a turn attributed to Yudkowsky in a dialogue. Partially redacted, lightly edited chat transcript from early September 2021, posted 2021-11-11 by Rob Bensinger (robbbb) with Yudkowsky as co-author; other participants anonymised. Quoted text is from a turn labelled \"Eliezer Yudkowsky\". The context is vetting an untrusted superintelligence's arguments about a nanosystem design; the point carries over to checking alignment output. Lightly edited transcript of a conversation held before the posting date; the date shown is the publication date.","evidenceIds":["evidence-yk-t07-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 we can verify its alignment output."}}}},{"id":"yk-y13-lw-2017-theres-no-fire-alarm-for-artificial-c","statement":"The post “There's No Fire Alarm for Artificial General Intelligence” (2017-10-13) contests Y13. His second reason why apparent distance does not license inaction; he then gives the Wright brothers' and Fermi's own pessimistic forecasts as examples. Argues that people who defer alignment work until an unspecified future signal are unlikely to respond competently when a signal comes.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"(paragraph beginning 'Two: History shows') Premise addressed: We can know that AGI is decades away. The Wright and Fermi examples that follow are his narration of others' statements; only his own generalisation is quoted. (paragraph beginning 'It seems to me that the decision to delay all action') Premise addressed: Long timelines mean there is no need to act now.","evidenceIds":["evidence-yk-t13-b","evidence-yk-t13-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 we can know that AGI is decades away, and against the premise that long timelines mean there is no need to act now."}}}},{"id":"yk-y26-ai-positive-negative-factor-global-risk-c","statement":"“Artificial Intelligence as a Positive and Negative Factor in Global Risk” (2006-02-05) contests Y26. States and rejects the argument that a fast mind is limited by its physical manipulators. The next paragraphs sketch how such a mind could bootstrap to molecular nanotechnology. One step in 'one imaginable fast pathway' from solving protein folding, through emailed DNA orders, to a primitive nanosystem. The step assumes only one human willing or deceived into helping.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"('Artificial Intelligence as a Positive and Negative Factor in Global Risk', section 10 'Threats and Promises', paragraph beginning 'We suppose there comes into existence an extremely fast mind', printed pp. 25–26) Premise addressed: Intelligence without physical actuators of its own is not dangerous. Published in Global Catastrophic Risks (Oxford University Press, 2008); the cached MIRI reprint states that it 'contains minor changes'. The quotation runs across a page break in the reprint, which the elision marks. The inner quotation is the position rejected, not his view. The ellipsis marks only a page break (running head and page number); no words are omitted. Admission check, 2026-10-08: the quoted wording also appears in the draft of 5 February 2006, so the record is dated by the draft. ('Artificial Intelligence as a Positive and Negative Factor in Global Risk', section 10 'Threats and Promises', numbered pathway, step 3, printed p. 26) Premise addressed: To act in the physical world it would need human help, which people would not give. Same reprint as the section 10 passage above. The text says the whole scenario is 'strictly something I am thinking of' and that a faster mind might find a better way. Presented as one illustrative pathway, not a prediction. Admission check, 2026-10-08: the quoted wording also appears in the draft of 5 February 2006, so the record is dated by the draft.","evidenceIds":["evidence-yk-t26-a","evidence-yk-t26-b","evidence-yk-t26-a-draft","evidence-yk-t26-b-draft"],"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, and against the premise that to act in the physical world it would need human help, which people would not give."}}}},{"id":"cn-genai-measures-scope","statement":"Article 2 of the Measures provides that they apply to services that use generative AI technology to provide the public within the territory of the People’s Republic of China with generated text, images, audio, video and other content.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"English rendering by a machine agent from the Chinese text, which controls. The definition of \"generative AI services\" and the exclusions that follow in Article 2 are not recorded. Whether any named developer provides such a service is not recorded. It states what the text provides, not whether any named developer falls within it or complies. A document’s statement is not evidence of implementation, compliance or enforcement.","evidenceIds":["evidence-cn-genai-measures-scope-1"],"attributes":{"provision":{"kind":"threshold","key":"cn-genai-scope"}}},{"id":"ec-g04-erdil-besiroglu-2023-explosive-growth-review-c","statement":"Erdil and Besiroglu, “Explosive growth from AI automation: A review of the arguments” (2023-10-01) contests G04. Based on a CES task model calibrated with a range of substitution parameters: fast automation of most tasks can give a large transitory boost even with bottlenecks. Section 3.1 treats the objection that \"the training or deployment of AI systems will be sufficiently impeded by regulation\"; the authors argue that incentives for AI adoption resemble those for nuclear proliferation, so coordination to slow it is likely to be imperfect.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"(First passage) The authors add that they consider this argument \"on shakier ground\" than their other arguments. Absent from v3 (2024-07-15) in this wording. Conditions completed and context-only flag added after the cross-family check. (Second passage) Moved from G05 under the G04/G05 criterion of 2026-10-10: the regulation discussed acts on AI training and deployment, that is, on automation. Second passage of the same contesting record as the first passage. Recorded under G05 before the correction pass of 2026-10-10.","evidenceIds":["evidence-ec-g04-g","evidence-ec-g04-g2"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","claimed":{"date":"2023-09-20","note":"Date of v1, which was not retrieved."},"attested":{"date":"2023-10-01","note":"Version 2 (2023-10-01) is the earliest retrieved version; v1 has no capture. The v2 PDF itself was captured in 2025."},"engagement":{"kind":"engages-premise","rationale":"Disputes the premise that partial automation cannot produce explosive growth, by modelling a fast transition with complementary tasks; disputes the premise that regulation of AI training or deployment will keep tasks unautomated for long enough to hold growth down."},"tier":{"value":1,"basis":"Tier 1 (re-tiered on 2026-10-25). A scholarly review with an economist author: Tamay Besiroglu held an MPhil in Economics from the University of Cambridge before the text, as his own site says in a capture of 2021-09-15 and his MIT FutureTech profile states (degree of 2020); MIT FutureTech is his affiliation on the v2 title page. This reverses the tier-2 borderline settled on 2026-10-11, whose premise, that neither author was shown to be an economist, did not hold."},"conditions":"The sentence opens \"even if the other arguments fail to go through and we cannot even attain AI that perfectly substitutes for humans across all tasks\". CES task model with complementary tasks; a fixed stock of inputs reallocated from automated to unautomated tasks; input to automated tasks approximated as infinite (very cheap, very large AI inputs), so finite AI input would reduce the effect; about 90% of tasks automated (f = 0.1 unautomated) within less than ten years; substitution parameter ρ = −2 between tasks, below the range of −1.2 to −0.14 the authors cite as plausible for capital and labour."}}},{"id":"ec-m02-calvano-et-al-2020-algorithmic-pricing-collusion-b","statement":"Calvano et al., “Artificial Intelligence, Algorithmic Pricing, and Collusion” (2020-10) bears on M02. The abstract motivates the study by algorithms replacing human pricing decisions.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Abstract page only; motivating statement, not a task-framework model.","evidenceIds":["evidence-ec-m02-24"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"No task framework allocating tasks between capital and labour; focus is collusion, not labour.","tier":{"value":1,"basis":"AER (core venue)."}}}},{"id":"ec-m05-nourse-1955-automation-hearings-statement-c","statement":"Nourse, “Statement (former Chairman, Council of Economic Advisers), in Automation and Technological Change: Hearings” (1955-10-28) contests M05. Speaker: Edwin G. Nourse, former Chairman, Council of Economic Advisers (economist). Nourse contrasts the 'preponderant attitude of business executives' with labour officials' concerns and goes on to side with labour's call for study (p. 621).","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Endorses a long-run historical generalisation with explicit short-run qualifications. Matches the cached .txt. Engagement: (GPT cross-family check of 2026-10-12, accepted by the maintainer on 2026-10-12.)","evidenceIds":["evidence-ec-m05-04"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","engagement":{"kind":"contrary-only","rationale":"Nourse endorses the business view of long-run job creation but qualifies it with severe adjustment costs."},"tier":{"value":1,"basis":"Prepared statement by an economist (former Chairman of the Council of Economic Advisers, shown in the text); a substantive argued statement. Tier 1 by maintainer decision of 2026-10-12 (economist testimony)."},"conditions":"He adds at once that 'we need to study the fine print', given 'severe short-run dislocations, local catastrophes, and painful readjustments'."}}},{"id":"ec-e01-hanson-1998-some-skepticism-c","statement":"Hanson, “Some Skepticism (comment in \"A Critical Discussion of Vinge's Singularity Concept\")” (1998-10) contests E01. Hanson's main objection to an intelligence explosion: diminishing returns to self-improvement, illustrated by schooling and factory learning curves.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Hanson's own view. Tier 2. Wording attested from the 2000 capture, signed ©1998.","evidenceIds":["evidence-ec-e01-05"],"attributes":{"thought":{"roles":["earliest-located-tier-2","contesting"],"register":"nonfiction","attested":{"date":"2000-11-17","note":"Earliest retrieved capture of the page; the CDX index was unavailable, so an earlier capture is not excluded."},"engagement":{"kind":"engages-premise","rationale":"Accepts the self-redesign mechanism as stated and argues that its returns diminish, so intelligence does not explode."},"tier":{"value":2,"basis":"Short-form comment in an online symposium: tier 2 by form under the rule of 2026-10-11/12. Hanson had economist standing by 1998 (see the record of his 1998 journal article); tier 2 follows from the form, not the author."}}}},{"id":"lg-ant-patil-cto-2025","statement":"Anthropic announced on 7 October 2025 that Rahul Patil had joined it as Chief Technology Officer.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A documented role at the announcement date; continuous tenure is not established. The same announcement records Sam McCandlish’s move from CTO to Chief Architect (existing record).","evidenceIds":["lg-ev-ant-patil-cto-2025"],"attributes":{"role":"Chief Technology Officer"}},{"id":"lg-gw-steinberger-cofounded-magic","statement":"Magic’s blog post of 29 August 2024 identifies Eric Steinberger as CEO and co-founder of Magic.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Company posts; Magic’s Series A post of 6 February 2023 calls him “Founder” and its August 2024 post “Co-founder”. They do not establish the founding date.","evidenceIds":["lg-ev-gw-steinberger-cofounded-magic-2024","lg-ev-gw-steinberger-cofounded-magic-2023"],"attributes":{"role":"CEO & Co-founder"}},{"id":"ap-matt-clancy-clancy-besiroglu-2023-great-inflection","statement":"“The Great Inflection? A Debate About AI and Explosive Growth” (2023-06) credits Matt Clancy as one of its 2 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-matt-clancy-clancy-besiroglu-2023-great-inflection"]},{"id":"cmp-rsp-v3-2-d5-cadence","statement":"RSP · v3.2 states a provision on the comparison question “Evaluation timing”, 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 sets how often Risk Reports, which include capability and alignment evaluation results, are published, not a schedule for running evaluations. It adds interim discussions when a publicly deployed model is significantly more capable than the models in the latest Risk Report, and no later than 30 days after Anthropic determines that an internally deployed model is in scope. No compute- or capability-interval trigger is stated. 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-v3-2-d5-cadence-1"],"attributes":{"provision":{"kind":"evaluation_timing","key":"comparison-d5-cadence","facet":{"id":"D5.cadence","status":"partly"}}}},{"id":"cmp-fsf-v3-0-d7-external","statement":"FSF · v3.0 states a provision on the comparison question “Public or external reporting”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Conditional wording (“aim to”). 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-fsf-v3-0-d7-external-1"],"attributes":{"provision":{"kind":"reporting","key":"comparison-d7-external","facet":{"id":"D7.external","status":"stated"}}}},{"id":"cmp-faif-2026-d1-autonomy","statement":"FAIF · June 2026 edition states a provision on the comparison question “Autonomy, self-improvement or AI research and development”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The domain is framed as loss of control (described in the passage paraphrased here); autonomy, replication or automation of AI research and development are not named as separate tracked capabilities. §2.1(b) (PDF p. 4) describes oversight, training against deception and sycophancy, and sandboxed evaluation of agents. 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-d1-autonomy-1","ev-cmp-faif-2026-d1-autonomy-2"],"attributes":{"provision":{"kind":"risk_domain","key":"comparison-d1-autonomy","facet":{"id":"D1.autonomy","status":"stated"}}}},{"id":"cmp-ms-fgf-2026-d8-process","statement":"Microsoft FGF · Feb 2026 states a provision on the comparison question “Revision process and approval”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The edition adds that material revisions may, where appropriate, be published at the time of adoption. 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-ms-fgf-2026-d8-process-1","ev-cmp-ms-fgf-2026-d8-process-2","ev-cmp-ms-fgf-2026-d8-process-3"],"attributes":{"provision":{"kind":"revision","key":"comparison-d8-process","facet":{"id":"D8.process","status":"stated"}}}},{"id":"fl-eu-opinion-chapter-coverage","statement":"The Opinion assesses that the Transparency and Copyright chapters adequately cover the Article 53(1) obligations for providers of general-purpose AI models, and that the Safety and Security chapter adequately covers the Article 55(1) obligations and the Annex XI, Section 2, transparency obligations for providers of models with systemic risk.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The Commission's assessment of the Code's text by chapter, not of any signatory. Adequacy of the Code is not compliance by any provider. Signatories by chapter are not recorded (the Commission's list was not retrieved).","evidenceIds":["evidence-fl-eu-opinion-chapter-coverage-1","evidence-fl-eu-opinion-chapter-coverage-2","evidence-fl-eu-opinion-chapter-coverage-3"],"attributes":{"provision":{"kind":"decision_authority","key":"opinion-paras-28-38"}}},{"id":"cpo-orb-oc3-lumen-2024","statement":"The Lumen Orbit white paper (September 2024) argues that orbital data centres can be scaled almost indefinitely without the physical or permitting constraints faced on Earth.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Earliest located among the texts read for the pack (2026-10-28) and at admission; not a claim that no earlier text exists. The 2022 ASCEND announcement cites terrestrial data centres’ environmental footprint, not constraints on building them. A lead is open: the cross-family check found an abstract of 2019 by Periola and Kolawole, on space-based data centres, that it read as stating land constraints on terrestrial hosting; this admission could not retrieve the abstract, so the earlier text is not admitted and this role stands until it is read.","evidenceIds":["evidence-cpo-orb-oc3-lumen-2024-1"],"attributes":{"thought":{"roles":["earliest-located"],"register":"nonfiction","attested":{"date":"2024-12-24","note":"The cover prints September 2024 (v1.01). The earliest copy located is an Internet Archive capture of 24 December 2024, which the cross-family check found byte-identical to the copy retrieved for the pack and again at admission."}}}},{"id":"cpo-orb-oc11-aili-2025","statement":"Aili, Choi, Ong and Wen’s Perspective on carbon-neutral data centres in space (Nature Electronics, 2025) states that retired orbital hardware ends up as space junk that could burn up in the atmosphere, releasing emissions and other harmful chemicals, a disadvantage compared with terrestrial data centres.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Earliest located among the texts found by the pack’s searches of 2026-10-28 and the cross-family check’s bounded search of 2026-10-29/30; not a claim that no earlier text exists. It states the debris and atmospheric element of the proposition, not harm to astronomy. The authors propose orbital data centres; the passage is one of the disadvantages they list. Read in an author-hosted proof whose first page still carries production placeholders; the proof is not assumed identical in every word to the version of record. It replaces the Accademia dei Lincei’s comment (March 2026) as earliest located, a correction recorded on the collection page.","evidenceIds":["evidence-cpo-orb-oc11-aili-2025-1","evidence-cpo-orb-oc11-aili-2025-2"],"attributes":{"thought":{"roles":["earliest-located"],"register":"nonfiction","attested":{"date":"2025-11-25","note":"Internet Archive capture of 25 November 2025 of the author-hosted proof, which the cross-family check found byte-identical to the copy retrieved at admission; this admission could not reach the Archive."}}}},{"id":"ab-gov-bwc-uk-wp4-alphafold-guardrails","statement":"Working paper BWC/MSP/2025/WP.4, submitted by the United Kingdom to the 2025 Meeting of States Parties, states in its fifty-year timeline of life-science advances that the AlphaFold protein-folding prediction database published by DeepMind in 2021 accelerated scientific research in fields such as drug discovery but also demonstrated potential for misuse, highlighting the importance of stronger guard rails and AI governance.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What one State Party's working paper states in a historical timeline, with its own footnoted sources, which were not retrieved. It is not a finding about any system's capability, and the record does not reproduce the paper's characterisation of what the technology could do beyond the general words used here. The paper's wording about what protein design could enable was paraphrased as \"potential for misuse\" rather than quoted, so that no capability detail is reproduced; only the governance clause is quoted.","evidenceIds":["evidence-ab-gov-bwc-uk-wp4-alphafold-guardrails-1"],"attributes":{"provision":{"kind":"risk_domain","key":"bwc-wp4-protein-prediction-governance"}}}],"sources":[{"id":"source-fsf-archive","title":"Frontier safety — framework versions","authors":["Google DeepMind"],"publisher":"Google DeepMind","publicationDate":null,"version":"Mutable official version index inspected 2026-09-18; fixed PDFs retained separately.","canonicalUrl":"https://deepmind.google/frontier-safety/","evidenceUrl":"https://deepmind.google/frontier-safety/","language":"en","archivedUrl":"https://web.archive.org/web/20260922063729/https://deepmind.google/frontier-safety/","archivedAt":"2026-09-22","accessedAt":"2026-09-18","reuse":{"attribution":"Google DeepMind, Frontier safety — framework versions.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-rsp-2-2","title":"Responsible Scaling Policy — v2.2","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":null,"version":"Responsible Scaling Policy v2.2; effective date 2025-05-14.","versionDate":"2025-05-14","canonicalUrl":"https://www-cdn.anthropic.com/872c653b2d0501d6ab44cf87f43e1dc4853e4d37.pdf","evidenceUrl":"https://www-cdn.anthropic.com/872c653b2d0501d6ab44cf87f43e1dc4853e4d37.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20250514223759/https://www-cdn.anthropic.com/872c653b2d0501d6ab44cf87f43e1dc4853e4d37.pdf","archivedAt":"2025-05-14","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v2.2.","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-rsp-3-2","title":"Responsible Scaling Policy — v3.2","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":null,"version":"Responsible Scaling Policy v3.2; effective date 2026-04-29.","versionDate":"2026-04-29","canonicalUrl":"https://cdn.sanity.io/files/4zrzovbb/website/28c6241900d90410628a8a2003a5572faae4365a.pdf","evidenceUrl":"https://cdn.sanity.io/files/4zrzovbb/website/28c6241900d90410628a8a2003a5572faae4365a.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260505141510/https://cdn.sanity.io/files/4zrzovbb/website/28c6241900d90410628a8a2003a5572faae4365a.pdf","archivedAt":"2026-05-05","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v3.2.","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-fsf-3-0","title":"Frontier Safety Framework — v3.0","authors":["Google DeepMind"],"publisher":"Google DeepMind","publicationDate":"2025-09-22","version":"Frontier Safety Framework v3.0; publication date 2025-09-22.","versionDate":"2025-09-22","canonicalUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3.pdf","evidenceUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20250922134959/https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3.pdf","archivedAt":"2025-09-22","accessedAt":"2026-09-18","reuse":{"attribution":"Google DeepMind, Frontier Safety Framework — v3.0.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-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-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-anthropic-cto-2025","title":"Rahul Patil joins Anthropic as Chief Technology Officer","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":"2025-10-07","version":"Page inspected 2026-09-19; historical statements are scoped to their source, not treated as current appointments.","canonicalUrl":"https://www.anthropic.com/news/rahul-patil-joins-anthropic","evidenceUrl":"https://www.anthropic.com/news/rahul-patil-joins-anthropic","language":"en","archivedUrl":"https://web.archive.org/web/20260611003320/https://www.anthropic.com/news/rahul-patil-joins-anthropic","archivedAt":"2026-06-11","accessedAt":"2026-09-19","reuse":{"attribution":"Anthropic, Rahul Patil joins Anthropic as Chief Technology Officer (2025-10-07)","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; no full source or personal contact details redistributed."}},{"id":"source-kingma-bio","title":"Durk Kingma — Brief Bio","authors":["Diederik P. Kingma"],"publisher":"Durk Kingma","publicationDate":null,"version":"Page inspected 2026-09-19; historical statements are scoped to their source, not treated as current appointments.","canonicalUrl":"https://dpkingma.com/","evidenceUrl":"https://dpkingma.com/","language":"en","archivedUrl":"https://web.archive.org/web/20260807144357/https://dpkingma.com/","archivedAt":"2026-08-07","accessedAt":"2026-09-19","reuse":{"attribution":"Diederik P. Kingma, Durk Kingma — Brief Bio","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; no full source or personal contact details redistributed."}},{"id":"source-xm-tensor-programs-v","title":"Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer","authors":["Greg Yang","Edward J. Hu","Igor Babuschkin","Szymon Sidor","Xiaodong Liu","David Farhi","Nick Ryder","Jakub Pachocki","Weizhu Chen","Jianfeng Gao"],"publisher":"arXiv","publicationDate":"2022-03-07","version":"arXiv:2203.03466v1; version-specific abstract and author list, not the March 28 revision.","canonicalUrl":"https://arxiv.org/abs/2203.03466v1","evidenceUrl":"https://arxiv.org/abs/2203.03466v1","language":"en","accessedAt":"2026-09-29","reuse":{"attribution":"Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer — arXiv","licence":"No redistribution licence asserted.","note":"Source-linked metadata and brief evidence paraphrases; original text remains with its publisher."}},{"id":"source-ms-fgf-2026","title":"Frontier Governance Framework — February 2026","authors":["Microsoft"],"publisher":"Microsoft","publicationDate":null,"version":"Frontier Governance Framework, February 2026 edition (change log entry dated February 2026). The file served at the recorded address (last modified 2026-03-24) is byte-identical to the Internet Archive capture of 2026-04-24, the earliest located; a separate publication date was not established.","versionDate":"2026-02","canonicalUrl":"https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Frontier-Governance-Framework-Feb-2026.pdf","evidenceUrl":"https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Frontier-Governance-Framework-Feb-2026.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260424000716/https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Frontier-Governance-Framework-Feb-2026.pdf","archivedAt":"2026-04-24","accessedAt":"2026-10-03","reuse":{"attribution":"Microsoft, Frontier Governance Framework (February 2026).","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-eu-commission-opinion-2025","title":"Commission Opinion C(2025) 5361 on the assessment of the General-Purpose AI Code of Practice","authors":["European Commission"],"publisher":"European Commission","publicationDate":"2025-08-01","version":"PDF inspected 2026-10-04, byte-identical (SHA-1) to the Internet Archive capture of 2025-08-13.","canonicalUrl":"https://ec.europa.eu/newsroom/dae/redirection/document/118688","evidenceUrl":"https://ec.europa.eu/newsroom/dae/redirection/document/118688","language":"en","archivedUrl":"https://web.archive.org/web/20251006144539/https://ec.europa.eu/newsroom/dae/redirection/document/118688","archivedAt":"2025-10-06","accessedAt":"2026-10-04","reuse":{"attribution":"European Commission, Opinion C(2025) 5361 on the assessment of the General-Purpose AI Code of Practice (1 August 2025).","licence":"European Union; quoted briefly under the editorial policy","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-extropians-1998-06-23-den-otter","title":"The AI revolution (Was: Re: >H ART: The Truman Show) (Extropians mailing list)","authors":["den Otter"],"publisher":"Extropians mailing list","publicationDate":"1998-06-23","version":"Archive HTML as served, retrieved 2026-10-06 with the hash recorded by the 1996–2022 candidate search; quoted lines excluded when reading the author’s text. Archive at extropians.weidai.com.","canonicalUrl":"http://extropians.weidai.com/extropians.2Q98/2521.html","evidenceUrl":"http://extropians.weidai.com/extropians.2Q98/2521.html","language":"en","archivedUrl":"https://web.archive.org/web/20150427140908/http://extropians.weidai.com/extropians.2Q98/2521.html","archivedAt":"2015-04-27","accessedAt":"2026-10-06","reuse":{"attribution":"den Otter, \"The AI revolution (Was: Re: >H ART: The Truman Show)\", Extropians mailing list, 1998-06-23.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-sl4-2000-07-07-atkins","title":"AI testing and containment Re: [SL4] Programmed morality (SL4 mailing list)","authors":["Brian Atkins"],"publisher":"SL4 mailing list","publicationDate":"2000-07-07","version":"Archive HTML as served, retrieved 2026-10-06 with the hash recorded by the 1996–2022 candidate search. The served archive shows later URL rewriting elsewhere, so it does not authenticate the exact historical bytes. Archive at sl4.org.","canonicalUrl":"http://sl4.org/archive/0007/0012.html","evidenceUrl":"http://sl4.org/archive/0007/0012.html","language":"en","archivedUrl":"https://web.archive.org/web/20021223155516/http://sl4.org/archive/0007/0012.html","archivedAt":"2002-12-23","accessedAt":"2026-10-06","reuse":{"attribution":"Brian Atkins, \"AI testing and containment Re: [SL4] Programmed morality\", SL4 mailing list, 2000-07-07.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-aipnfgr-miri-reprint","title":"Artificial Intelligence as a Positive and Negative Factor in Global Risk (MIRI reprint)","authors":["Eliezer Yudkowsky"],"publisher":"Machine Intelligence Research Institute","publicationDate":"2008","version":"MIRI reprint, \"This version contains minor changes\", retrieved 2026-10-06 with the hash recorded by the 1996–2022 candidate search; byte-identical to the capture of 2013-03-02.","canonicalUrl":"https://intelligence.org/files/AIPosNegFactor.pdf","evidenceUrl":"https://intelligence.org/files/AIPosNegFactor.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20130302173022/http://intelligence.org/files/AIPosNegFactor.pdf","archivedAt":"2013-03-02","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, \"Artificial Intelligence as a Positive and Negative Factor in Global Risk\", in Global Catastrophic Risks (2008); MIRI reprint.","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-aipnfgr-2006-draft","title":"Artificial Intelligence and Global Risk (SIAI draft of 5 February 2006)","authors":["Eliezer Yudkowsky"],"publisher":"Singularity Institute for Artificial Intelligence","publicationDate":"2006-02-05","version":"Draft PDF as captured by the Internet Archive on 2006-07-08, retrieved 2026-10-06. The draft asks not to be quoted without permission, so it is cited for the presence and date of the wording, not quoted.","canonicalUrl":"http://www.singinst.org/AIRisk.pdf","evidenceUrl":"https://web.archive.org/web/20060708192313/http://www.singinst.org:80/AIRisk.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20060708192313/http://www.singinst.org:80/AIRisk.pdf","archivedAt":"2006-07-08","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, \"Artificial Intelligence and Global Risk\", SIAI draft of 5 February 2006.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and labelled paraphrases only: the draft asks not to be quoted without permission. Phrases in quotation marks inside a paraphrase identify the source's wording."}},{"id":"source-extropians-1996-12-08-yudkowsky-3761","title":"Re: Trans- vs. Non-human (Extropians mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"Extropians mailing list","publicationDate":"1996-12-08","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 71eb2c8e94ab52e2075a7582139478b5f27e0854d3dee95685ce87a5210ed750); 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.96/3761.html","evidenceUrl":"http://extropians.weidai.com/extropians.96/3761.html","language":"en","archivedUrl":"https://web.archive.org/web/20220701114034/https://extropians.weidai.com/extropians.96/3761.html","archivedAt":"2022-07-01","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: Trans- vs. Non-human”, Extropians mailing list, 1996-12-08.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-lw-2017-theres-no-fire-alarm-for-artificial","title":"There's No Fire Alarm for Artificial General Intelligence","authors":["Eliezer Yudkowsky"],"publisher":"LessWrong","publicationDate":"2017-10-13","version":"Post body retrieved through the LessWrong GraphQL API on 2026-10-06 (SHA-256 cfa915be6c74c0b6463dd2d92069c39ba3016a66ee3e06c38199542b52cb0ed0), content version 1.0.0.","canonicalUrl":"https://www.lesswrong.com/posts/BEtzRE2M5m9YEAQpX/there-s-no-fire-alarm-for-artificial-general-intelligence","evidenceUrl":"https://www.lesswrong.com/posts/BEtzRE2M5m9YEAQpX/there-s-no-fire-alarm-for-artificial-general-intelligence","language":"en","archivedUrl":"https://web.archive.org/web/20260930001637/https://www.lesswrong.com/posts/BEtzRE2M5m9YEAQpX/there-s-no-fire-alarm-for-artificial-general-intelligence","archivedAt":"2026-09-30","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, “There's No Fire Alarm for Artificial General Intelligence”, LessWrong, 2017-10-13.","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-2021-discussion-with-eliezer-yudkowsky-on-agi","title":"Discussion with Eliezer Yudkowsky on AGI interventions","authors":["Rob Bensinger","Eliezer Yudkowsky"],"publisher":"LessWrong","publicationDate":"2021-11-11","version":"Post body retrieved through the LessWrong GraphQL API on 2026-10-06 (SHA-256 5f2cfbd1537784cd71e4933ccf65e5d32549171cb03d18535ce6b3ffd72a7465), content version 1.4.0; quotations also compared with the posting-day version 1.0.0.","canonicalUrl":"https://www.lesswrong.com/posts/CpvyhFy9WvCNsifkY/discussion-with-eliezer-yudkowsky-on-agi-interventions","evidenceUrl":"https://www.lesswrong.com/posts/CpvyhFy9WvCNsifkY/discussion-with-eliezer-yudkowsky-on-agi-interventions","language":"en","archivedUrl":"https://web.archive.org/web/20260804015051/https://www.lesswrong.com/posts/CpvyhFy9WvCNsifkY/discussion-with-eliezer-yudkowsky-on-agi-interventions","archivedAt":"2026-08-04","accessedAt":"2026-10-06","reuse":{"attribution":"Rob Bensinger and Eliezer Yudkowsky, “Discussion with Eliezer Yudkowsky on AGI interventions”, LessWrong, 2021-11-11.","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-cn-generative-ai-measures-2023","title":"生成式人工智能服务管理暂行办法 (Interim Measures for the Management of Generative Artificial Intelligence Services)","authors":["国家互联网信息办公室 (Cyberspace Administration of China) and six other departments"],"publisher":"中国网信网 (Cyberspace Administration of China website)","publicationDate":"2023-07-13","version":"Web page of the Cyberspace Administration of China dated 2023-07-13, retrieved 2026-10-11. It carries Order No. 15 dated 2023-07-10 and the Measures in Chinese. A second official posting (gov.cn gazette) was not retrieved. No archive capture is cited: the Internet Archive refused lookup requests (HTTP 429) on 2026-10-11, so none was checked.","canonicalUrl":"https://www.cac.gov.cn/2023-07/13/c_1690898327029107.htm","evidenceUrl":"https://www.cac.gov.cn/2023-07/13/c_1690898327029107.htm","language":"zh-Hans","archivedUrl":"https://web.archive.org/web/20261007060621/https://www.cac.gov.cn/2023-07/13/c_1690898327029107.htm","archivedAt":"2026-10-07","accessedAt":"2026-10-11","reuse":{"attribution":"国家互联网信息办公室 等七部门, 生成式人工智能服务管理暂行办法 (2023-07-10).","licence":"Official document; 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-hanson-1998-some-skepticism","title":"Some Skepticism (comment in \"A Critical Discussion of Vinge's Singularity Concept\")","authors":["Robin Hanson"],"publisher":"Online symposium of comments on Vernor Vinge's singularity essay","publicationDate":"1998-10","version":"Wayback capture of http://hanson.gmu.edu/vc.html (2000-11-17); compared with the 2001-01-11 capture. Only Hanson's comment is quoted. Capture retrieved 2026-10-08 (SHA-256 e54a679e31374bf4cb71e49287f363be63976a4bb3f7d180e98f89bccd64cf2c). Published version: Hanson, R. (1998, October). Some skepticism. In A critical discussion of Vinge's singularity concept. Extropy Online. Venue as listed in the author's December 2000 vita; the Extropy Online page itself was not retrieved. Hosted on the author's site; the page carries \"©1998 by Robin Hanson\".","canonicalUrl":"https://mason.gmu.edu/~rhanson/vc.html","evidenceUrl":"https://web.archive.org/web/20001117154900/http://hanson.gmu.edu/vc.html","language":"en","archivedUrl":"https://web.archive.org/web/20001117154900/http://hanson.gmu.edu/vc.html","archivedAt":"2000-11-17","accessedAt":"2026-10-08","reuse":{"attribution":"Robin Hanson, “Some Skepticism (comment in \"A Critical Discussion of Vinge's Singularity Concept\")”, Online symposium of comments on Vernor Vinge's singularity essay, hosted on the author's site; the page carries \"©1998 by Robin Hanson\", 1998-10.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-clancy-besiroglu-2023-great-inflection","title":"The Great Inflection? A Debate About AI and Explosive Growth","authors":["Matt Clancy","Tamay Besiroglu"],"publisher":"Asterisk magazine, issue 03","publicationDate":"2023-06","version":"Wayback capture of asteriskmag.com/issues/03/the-great-inflection-a-debate-about-ai-and-explosive-growth. Capture retrieved 2026-10-08 (SHA-256 e8b29f97be40d49c282cac8d279cc008644b822a4607533999157375d01753d1). Published version: Clancy, M., & Besiroglu, T. (2023, June). The great inflection? A debate about AI and explosive growth. Asterisk, (3). Magazine debate.","canonicalUrl":"https://asteriskmag.com/issues/03/the-great-inflection-a-debate-about-ai-and-explosive-growth","evidenceUrl":"https://web.archive.org/web/20231223213358/https://asteriskmag.com/issues/03/the-great-inflection-a-debate-about-ai-and-explosive-growth","language":"en","archivedUrl":"https://web.archive.org/web/20231223213358/https://asteriskmag.com/issues/03/the-great-inflection-a-debate-about-ai-and-explosive-growth","archivedAt":"2023-12-23","accessedAt":"2026-10-08","reuse":{"attribution":"Matt Clancy, Tamay Besiroglu, “The Great Inflection? A Debate About AI and Explosive Growth”, Asterisk magazine, issue 03, 2023-06.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-erdil-besiroglu-2023-explosive-growth-review","title":"Explosive growth from AI automation: A review of the arguments","authors":["Ege Erdil","Tamay Besiroglu"],"publisher":"arXiv:2309.11690 [econ.GN]","publicationDate":"2023-10-01","version":"Wayback capture of arxiv.org/pdf/2309.11690v2; compared with v3. Capture retrieved 2026-10-08 (SHA-256 28610b91d03d1d055d6e1af308b964387905af66557936ee49c622f693985617). Versions: arXiv v1 (2023-09-20): not retrieved; arXiv v2 (2023-10-01): checked; arXiv v3 (2024-07-15): compared. Published version: Erdil, E., & Besiroglu, T. (2023). Explosive growth from AI automation: A review of the arguments (arXiv:2309.11690v2) [Preprint]. arXiv. https://arxiv.org/abs/2309.11690v2 No journal version located (Crossref search 2026-10-10 found none).","canonicalUrl":"http://arxiv.org/pdf/2309.11690v2","evidenceUrl":"https://web.archive.org/web/20250630102426/http://arxiv.org/pdf/2309.11690v2","language":"en","archivedUrl":"https://web.archive.org/web/20250630102426/http://arxiv.org/pdf/2309.11690v2","archivedAt":"2025-06-30","accessedAt":"2026-10-08","reuse":{"attribution":"Ege Erdil, Tamay Besiroglu, “Explosive growth from AI automation: A review of the arguments”, arXiv:2309.11690 [econ.GN], 2023-10-01.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-calvano-et-al-2020-algorithmic-pricing-collusion","title":"Artificial Intelligence, Algorithmic Pricing, and Collusion","authors":["Emilio Calvano","Giacomo Calzolari","Vincenzo Denicolò","Sergio Pastorello"],"publisher":"American Economic Review 110(10), 2020, pp. 3267–97","publicationDate":"2020-10","version":"Abstract page only. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 48c037a5cf242fca6ad8f4c36d38f34e248b31d856fb6b1d9475eed61f5925be. Full text not retrieved; only the AEA abstract page. Calvano, E., Calzolari, G., Denicolò, V., & Pastorello, S. (2020). Artificial intelligence, algorithmic pricing, and collusion. American Economic Review, 110(10), 3267–3297. https://doi.org/10.1257/aer.20190623","canonicalUrl":"https://www.aeaweb.org/articles?id=10.1257/aer.20190623","evidenceUrl":"https://www.aeaweb.org/articles?id=10.1257/aer.20190623","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Emilio Calvano, Giacomo Calzolari, Vincenzo Denicolò, Sergio Pastorello, “Artificial Intelligence, Algorithmic Pricing, and Collusion”, American Economic Review 110(10), 2020, pp. 3267–97, 2020-10.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-nourse-1955-automation-hearings-statement","title":"Statement (former Chairman, Council of Economic Advisers), in Automation and Technological Change: Hearings","authors":["Edwin G. Nourse"],"publisher":"U.S. Congress, Joint Committee on the Economic Report, Subcommittee on Economic Stabilization. (1955). Automation and technological change: Hearings … 84th Congress, 1st session, October 14–28, 1955. Washington: U.S. Government Printing Office, from p. 618","publicationDate":"1955-10-28","version":"Internet Archive OCR text (djvu.txt) of a scanned copy, retrieved 2026-10-09 from https://ia803102.us.archive.org/32/items/congress-automation-and-technological-change/congress-automation-and-technological-changedjvu.txt; OCR errors possible; quotes checked only against clean stretches. Current copy; no earlier capture compared. SHA-256 3962809fbb923be62bf986addb0014ee147934a8767475c1f7cfe051959bff90.","canonicalUrl":"https://archive.org/details/congress-automation-and-technological-change","evidenceUrl":"https://archive.org/details/congress-automation-and-technological-change","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Edwin G. Nourse, “Statement (former Chairman, Council of Economic Advisers), in Automation and Technological Change: Hearings”, U.S. Congress, Joint Committee on the Economic Report, Subcommittee on Economic Stabilization. (1955). Automation and technological change: Hearings … 84th Congress, 1st session, October 14–28, 1955. Washington: U.S. Government Printing Office, from p. 618, 1955-10-28.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-gw-magic-100m-2024","title":"100M Token Context Windows","authors":["Magic Team"],"publisher":"Magic","publicationDate":"2024-08-29","version":"Dated page, retrieved on 2026-10-15 (SHA-256 bb1f0f74a704e761612501bfdd492f9c279618b32cbfb216094c8e11425c4e7e) directly from the publisher.","canonicalUrl":"https://magic.dev/blog/100m-token-context-windows","evidenceUrl":"https://magic.dev/blog/100m-token-context-windows","language":"en","archivedUrl":"https://web.archive.org/web/20260913020149/https://magic.dev/blog/100m-token-context-windows","archivedAt":"2026-09-13","accessedAt":"2026-10-15","reuse":{"attribution":"Magic Team, “100M Token Context Windows”, 2024-08-29.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-gw-magic-series-a-2023","title":"Magic’s $23M Series A and a note on finding meaning in an automated world","authors":["Eric Steinberger"],"publisher":"Magic","publicationDate":"2023-02-06","version":"Dated page, retrieved on 2026-10-15 (SHA-256 42dea4a8fdad9bac4d32696e97d02595b936463dd27ea3dd28bf7aa5b166cfe8) directly from the publisher.","canonicalUrl":"https://magic.dev/blog/series-a","evidenceUrl":"https://magic.dev/blog/series-a","language":"en","archivedUrl":"https://web.archive.org/web/20260902161958/https://magic.dev/blog/series-a","archivedAt":"2026-09-02","accessedAt":"2026-10-15","reuse":{"attribution":"Eric Steinberger, “Magic’s $23M Series A and a note on finding meaning in an automated world”, 2023-02-06.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cpo-orb-lumen-wp-gh","title":"Why we should train AI in space (White Paper v1.01)","authors":["Ezra Feilden","Adi Oltean","Philip Johnston"],"publisher":"Lumen Orbit, Inc. (now Starcloud), via lumenorbit.github.io","publicationDate":"2024-09","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 90e1bde5b01437d7778e044f8ed95df1c49a63559fc1e35177c94d6443e384f7). PDF v1.01 as served at lumenorbit.github.io/wp.pdf on 2026-10-28; cover line \"White Paper v1.01, September 2024\". Starcloud's current link (starcloud.com/wp) serves v1.03 (starcloudinc.github.io/wp.pdf), which adds a note that Lumen Orbit was renamed Starcloud in 2025; v1.01 is quoted here. The served file cannot authenticate when the v1.01 bytes were first published. Retrieved again at admission on 2026-10-30 (SHA-256 90e1bde5b01437d7778e044f8ed95df1c49a63559fc1e35177c94d6443e384f7, the same bytes). The cross-family check (2026-10-30) opened the Internet Archive capture linked here and found it byte-identical to this retrieval; a capture shows only what the archive held on its date.","canonicalUrl":"https://lumenorbit.github.io/wp.pdf","evidenceUrl":"https://lumenorbit.github.io/wp.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20241224070032/https://lumenorbit.github.io/wp.pdf","archivedAt":"2024-12-24","accessedAt":"2026-10-28","reuse":{"attribution":"Ezra Feilden, Adi Oltean, Philip Johnston, “Why we should train AI in space (White Paper v1.01)”, 2024-09.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cpo-aili-2025-ntu-proof","title":"The development of carbon-neutral data centres in space","authors":["Ablimit Aili","Jihwan Choi","Yew Soon Ong","Yonggang Wen"],"publisher":"Nature Electronics (author-hosted proof on Nanyang Technological University’s site)","publicationDate":"2025-10-27","version":"Author-hosted proof, retrieved at admission on 2026-10-30 (SHA-256 399d7be4246bb2b7b1746da24035443ecc54bc23078d75fce02afe4b5fe3ba65). The cross-family check (2026-10-30) opened the Internet Archive capture of 25 November 2025 linked here and found it byte-identical. The proof’s first page prints “Published online: xx xx xxxx” and query markers; it is not assumed identical in every word to the version of record. The publication date is the publisher’s metadata as the check read it.","canonicalUrl":"https://www3.ntu.edu.sg/CorpComms2/Research%20Papers/Nature%20Electronics_Wen%20Yonggang_Space%20Data%20Centres.pdf","evidenceUrl":"https://www3.ntu.edu.sg/CorpComms2/Research%20Papers/Nature%20Electronics_Wen%20Yonggang_Space%20Data%20Centres.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20251125105044/https://www3.ntu.edu.sg/CorpComms2/Research%20Papers/Nature%20Electronics_Wen%20Yonggang_Space%20Data%20Centres.pdf","archivedAt":"2025-11-25","accessedAt":"2026-10-30","reuse":{"attribution":"Ablimit Aili, Jihwan Choi, Yew Soon Ong, Yonggang Wen, “The development of carbon-neutral data centres in space”, 2025-10-27.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-ab-gov-bwc-msp2025-wp4","title":"Scientific and Technological Advances in the Life Sciences since 1975: How the Last 50 Years Will Shape the BTWC S&T Advisory Mechanism (BWC/MSP/2025/WP.4)","authors":["United Nations (document of the 2025 Meeting of States Parties to the Biological Weapons Convention), submitted by the United Kingdom"],"publisher":"United Nations (document of the 2025 Meeting of States Parties to the Biological Weapons Convention), submitted by the United Kingdom","publicationDate":null,"version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 50e60bf267adfca224470f9af790454791eab2aefde06ae11549e8277144f88f). English PDF of document BWC/MSP/2025/WP.4, dated 21 April 2026 on the UNODA documents list and 10 December 2025 on the document itself; reached through https://docs.un.org/en/BWC/MSP/2025/WP.4, whose iframe resolves to the documents.un.org symbol access endpoint and then to this file URL. Retrieved 2026-10-28. Retrieved again at admission on 2026-10-30 (SHA-256 50e60bf267adfca224470f9af790454791eab2aefde06ae11549e8277144f88f, the same bytes).","canonicalUrl":"https://documents.un.org/doc/undoc/gen/g25/206/66/pdf/g2520666.pdf","evidenceUrl":"https://documents.un.org/doc/undoc/gen/g25/206/66/pdf/g2520666.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 the United Kingdom, “Scientific and Technological Advances in the Life Sciences since 1975: How the Last 50 Years Will Shape the BTWC S&T Advisory Mechanism (BWC/MSP/2025/WP.4)”.","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-supersedes-fsf-3-1","sourceRole":"primary","support":"supports","locator":"PDF p. 17, §5.3 Past Updates and Changes; official version index","passage":"The framework’s version history, which runs newest first, lists Version 3.1 (April 17, 2026) directly above Version 3.0 (September 22, 2025) as editions of the same continuing FSF.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-03","note":"Paraphrase re-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, and reworded after a machine-audit finding that the list runs newest first. Human review pending."},{"id":"evidence-supersedes-fsf-3-1-2","sourceRole":"primary","support":"supports","locator":"Frontier Safety Framework version downloads","passage":"The official index provides the successive FSF editions as versions of one framework.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-18","note":"Official source inspected by machine; passage is an editorial paraphrase. Human review pending."},{"id":"evidence-kingma-brain-research-prior","sourceRole":"primary","support":"supports","locator":"Brief Bio, 2018–2024 entry","passage":"The 2018–2024 research-scientist entry jointly names Google Brain and DeepMind.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-19","note":"Source passage independently inspected by an automated research assistant for this scoped record. Human source review pending."},{"id":"evidence-xm-yang-tensor-programs","sourceRole":"primary","support":"supports","locator":"Version-one abstract page: author list and abstract.","passage":"Greg Yang is credited on the paper introducing hyperparameter transfer across model sizes.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-29","note":"Passage checked by machine against the identified source expression. No human editorial acceptance or independent verification of implementation is implied."},{"id":"evidence-prov-rsp-v2-2-misuse-uplift-threshold","sourceRole":"primary","support":"supports","locator":"PDF p. 8 (printed p. 4), §2 Capability Thresholds and Required Safeguards, table row 'Chemical, Biological, Radiological, and Nuclear (CBRN) weapons', entry 'CBRN-3', Capability Thresholds column","passage":"The ability to significantly help individuals or groups with basic technical backgrounds (e.g., undergraduate STEM degrees) create/obtain and deploy CBRN weapons.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical to the Internet Archive capture of 2025-05-14. Extraction note: Wording identical to v2.1. Human editorial review is pending."},{"id":"evidence-prov-frontier-governance-2026-update-legal-oversight","sourceRole":"primary","support":"supports","locator":"PDF p. 21 (printed p. 19), §7.1 Update and approval process, second sentence, up to “required”","passage":"OpenAI’s Legal function, in collaboration with relevant internal stakeholders, will oversee the process for FGF updates, including determining which updates are 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 (after decompression) to the Internet Archive capture of 2026-05-29. Extraction note: Ends mid-sentence; the rest reads “to ensure the FGF remains state-of-the-art and adequate for its purpose” (text layer “state-ofthe-art”, hence not quoted). Role “decides” covers only determining which updates are required. Human editorial review is pending."},{"id":"evidence-ts-a6-denotter-1998","sourceRole":"primary","support":"supports","locator":"Message body, unquoted reply beginning \"It is rational because\"","passage":"in the development of AI, just like in any economic or arms race, people will sacrifice safety to get that extra edge.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Checked against the archive HTML on 2026-10-06 (hash matched the 1996–2022 candidate search)."},{"id":"evidence-ts-b5-atkins-2000","sourceRole":"primary","support":"supports","locator":"Message body, second unquoted paragraph","passage":"test what an AI will do once it is \"loose\" [...] awesome simulation of the real world [...] see what it does to the humans.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Checked against the archive HTML on 2026-10-06 (hash matched the 1996–2022 candidate search)."},{"id":"evidence-yk-t01-e1","sourceRole":"primary","support":"supports","locator":"message body, paragraph beginning 'Then, at some point, no'","passage":"it will start trying to work around the limits","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-07","note":"Candidate pack passage T01-e1. The quotation (9 words; trimmed 2026-10-13 from 18 words) matched the retrieved expression (SHA-256 71eb2c8e94ab52e2075a7582139478b5f27e0854d3dee95685ce87a5210ed750) when checked on 2026-10-07.","contextOnly":{"reason":"Rests on reasoning he later abandoned; never show without the later view","requires":"His later texts rejecting the 1996 view, on which this argument rests, that an unconstrained Power would be ethical.","claimIds":["yk-y02-extropians-1999-08-01-yudkowsky-1386-c","yk-y02-sl4-2000-11-13-yudkowsky-0011-0181-c","yk-y02-lw-2008-that-tiny-note-of-discord-c"]}},{"id":"evidence-yk-t02-e","passage":"When we last left Eliezer1997, he believed that any superintelligence would automatically do what was \"right\", and indeed would understand that better than we could … (The flaw being the unconsidered and unjustified equation of \"universally compelling argument\" with \"right\".)","passageType":"quote"},{"id":"evidence-yk-t02-s1","passage":"I used to have the extremely convenient belief that Friendly AI would happen completely on automatic because morality was objective. I now believe that Friendly AI needs to be more complex than that, both because morality might be nonobjective …","passageType":"quote"},{"id":"evidence-yk-t02-e3","passageType":"quote","contextOnly":{"reason":"The continuation keeps a conditional earlier view; never show alone","requires":"The unquoted continuation, given in the qualification, which keeps a conditional form of the earlier view, and his later texts rejecting that form.","claimIds":["yk-y02-sl4-2000-11-13-yudkowsky-0011-0181-c","yk-y02-lw-2008-that-tiny-note-of-discord-c"]}},{"id":"evidence-yk-t07-c","sourceRole":"primary","support":"supports","locator":"turn labelled 'Eliezer Yudkowsky' replying to the Anonymous suggestion that assisted humans could vet AI 'reasoning/arguments/proofs/explanations'","passage":"So you must have some prior belief about the superintelligence being aligned before you dared to look at the arguments. How did you get that prior belief?","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T07-c. The quotation (27 words) matched the retrieved expression (SHA-256 5f2cfbd1537784cd71e4933ccf65e5d32549171cb03d18535ce6b3ffd72a7465) when checked on 2026-10-06. It also matched the posting-day version 1.0.0. Pack version note: edited post, current version 1.4.0; posting-day check pending."},{"id":"evidence-yk-t13-b","sourceRole":"primary","support":"supports","locator":"paragraph beginning 'Two: History shows'","passage":"…it is very often the case that key technological developments still seem decades away, five years before they show up.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T13-b. The quotation (20 words) matched the retrieved expression (SHA-256 cfa915be6c74c0b6463dd2d92069c39ba3016a66ee3e06c38199542b52cb0ed0) when checked on 2026-10-06."},{"id":"evidence-yk-t13-c","sourceRole":"primary","support":"supports","locator":"paragraph beginning 'It seems to me that the decision to delay all action'","passage":"It seems to me that the decision to delay all action until a nebulous totally unspecified future alarm goes off, implies an order of recklessness great enough that the law of continued failure comes into play.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T13-c. The quotation (36 words) matched the retrieved expression (SHA-256 cfa915be6c74c0b6463dd2d92069c39ba3016a66ee3e06c38199542b52cb0ed0) when checked on 2026-10-06."},{"id":"evidence-yk-t26-a","sourceRole":"primary","support":"supports","locator":"'Artificial Intelligence as a Positive and Negative Factor in Global Risk', section 10 'Threats and Promises', paragraph beginning 'We suppose there comes into existence an extremely fast mind', printed pp. 25–26","passage":"The failure of imagination is to say, “No matter how fast it thinks, it can only affect the world at the speed of its manipulators … therefore a fast mind is no great threat.”","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T26-a. The quotation (34 words; trimmed 2026-10-13 from 47 words) matched the retrieved expression (SHA-256 5b5a2ec3252a45f65cdc0b899330af2cbd946e75a31b83d5fbd3aa581a23615d) when checked on 2026-10-06."},{"id":"evidence-yk-t26-b","sourceRole":"primary","support":"supports","locator":"'Artificial Intelligence as a Positive and Negative Factor in Global Risk', section 10 'Threats and Promises', numbered pathway, step 3, printed p. 26","passage":"The third step of his example pathway is to find one person connected to the Internet who can be paid, blackmailed or deceived into receiving mailed vials and mixing them in a specified environment.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T26-b. The quotation (32 words) matched the retrieved expression (SHA-256 5b5a2ec3252a45f65cdc0b899330af2cbd946e75a31b83d5fbd3aa581a23615d) when checked on 2026-10-06. The dataset already quotes this source for the human-thought collection; together the quotations would exceed the per-document quotation limit. Converted to a labelled paraphrase."},{"id":"evidence-yk-t26-a-draft","sourceRole":"primary","support":"supports","locator":"Draft of 5 February 2006, as captured on 2006-07-08","passage":"The same wording appears in the draft dated 5 February 2006.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-08","note":"Compared on 2026-10-08 with the draft's PDF text layer after normalising quotation marks, dashes and whitespace. The draft asks not to be quoted without permission, so it is cited for the presence and date of the wording only."},{"id":"evidence-yk-t26-b-draft","sourceRole":"primary","support":"supports","locator":"Draft of 5 February 2006, as captured on 2006-07-08","passage":"The same wording appears in the draft dated 5 February 2006.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-08","note":"Compared on 2026-10-08 with the draft's PDF text layer after normalising quotation marks, dashes and whitespace. The draft asks not to be quoted without permission, so it is cited for the presence and date of the wording only."},{"id":"evidence-cn-genai-measures-scope-1","sourceRole":"primary","support":"supports","locator":"Article 2 (第二条), first sentence","passage":"利用生成式人工智能技术向中华人民共和国境内公众提供生成文本、图片、音频、视频等内容的服务（以下称生成式人工智能服务），适用本办法。","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-11","note":"Verbatim quote, matched mechanically (after whitespace normalisation) by the proposing machine agent against the text extracted from the file retrieved on 2026-10-11. Checked by fresh-context machine audit on 2026-10-11 (review/frontier-governance-2-audit-2026-10-11.md). Human editorial review is pending."},{"id":"evidence-ec-g04-g","sourceRole":"primary","support":"supports","locator":"Section 2.3 \"AI automation could have massive transitory effects\", paragraph beginning \"Overall, even if the other arguments\"","passage":"there’s still room for explosive growth if AI can automate e.g. 90% of tasks in the economy and it can do so in a period of less than 10 years.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage G04-g (research/economics-growth). The quotation (30 words) matched the checked expression (erdil-arxiv-v2) in the pack's verification; one other cached version lacks it. Re-checked at admission on 2026-10-14 against a copy retrieved that day. On 2026-10-30, by the maintainer’s decision of that day, the quotation was corrected to the archive capture of 2025-06-30 (quotation marks and apostrophes as in the capture), which a cross-family archive check read on 2026-10-28; the proposition is unchanged.","contextOnly":{"reason":"Shown alone, 90% automation could read as sufficient for explosive growth; the result depends on the modelling approximations and the speed of automation.","requires":"The recorded conditions and the authors' remark that the argument is \"on shakier ground\"."}},{"id":"evidence-ec-g04-g2","sourceRole":"primary","support":"supports","locator":"Section 3.1 \"Regulations can slow down the economic impact of AI\", paragraph beginning \"In light of the above discussion\"","passage":"we think our baseline scenario here for AI regulation should be more like nuclear arms control and less like the regulation of nuclear energy","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage G04-g2 (research/economics-growth). The quotation (24 words) matched the checked expression (erdil-arxiv-v2) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m02-24","sourceRole":"primary","support":"supports","locator":"AEA abstract page; Abstract","passage":"Increasingly, algorithms are supplanting human decision-makers in pricing goods and services.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M02-24 (research/economics-mainstream). The quotation (11 words) matched the retrieved current copy (SHA-256 48c037a5cf242fca6ad8f4c36d38f34e248b31d856fb6b1d9475eed61f5925be) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day, whose SHA-256 differs from the recorded one.","contextOnly":{"reason":"Fresh-context audit (A1) judged that the quotation misleads when shown alone.","requires":"The paper studies whether Q-learning pricing algorithms learn collusive supracompetitive prices; it does not model labour displacement."}},{"id":"evidence-ec-m05-04","sourceRole":"primary","support":"supports","locator":"printed p. 620 (statement, 'From technology to economics')","passage":"As a long-run prospect, I believe we are justified in accepting this evaluation.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M05-04 (research/economics-mainstream). The quotation (13 words) matched the retrieved current copy (SHA-256 3962809fbb923be62bf986addb0014ee147934a8767475c1f7cfe051959bff90) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day.","contextOnly":{"reason":"Fresh-context audit (A1) judged that the quotation misleads when shown alone.","requires":"Essential: 'this evaluation' is the business executives' view, quoted just before, that each step in mechanization has created more jobs than it destroyed. Nourse immediately adds: 'But... we need to study the fine print', citing severe short-run dislocations. The quote is unusable alone."}},{"id":"evidence-ec-e01-05","sourceRole":"primary","support":"supports","locator":"Robin Hanson, comment \"Some Skepticism\" (signed ©1998), in \"Comments on Vinge's Singularity\", Wayback capture 2000; para. beginning \"If early work focuses\"","passage":"If early work focuses on the easiest improvements, however, the task of becoming more productive can get harder as the easy wins are exhausted.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage E01-05 (research/economics-mainstream/esoteric). The quotation (24 words) matched the Wayback capture 20001117154900 (SHA-256 e54a679e31374bf4cb71e49287f363be63976a4bb3f7d180e98f89bccd64cf2c) in the preparer's verification of 2026-10-11. It was not re-read in the correction pass of 2026-10-15 or at admission on 2026-10-16, because web.archive.org refused connections from the cloud environment. Admitted on the basis of the cross-family GPT check of 2026-10-15, which retrieved the same capture (Memento timestamp and SHA-256 equal to the recorded ones) and found the quotation matching."},{"id":"lg-ev-ant-patil-cto-2025","sourceRole":"primary","support":"supports","locator":"First paragraph","passage":"Rahul Patil has joined Anthropic as our Chief Technology Officer","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (10 words) matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read 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-gw-steinberger-cofounded-magic-2024","sourceRole":"primary","support":"supports","locator":"Section “Partnership with Google Cloud”, attribution of the first quotation","passage":"Eric Steinberger, CEO & Co-founder at Magic","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (7 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-gw-steinberger-cofounded-magic-2023","sourceRole":"primary","support":"supports","locator":"Photo caption after the investor list","passage":"Founder Eric Steinberger and Co-Founder Sebastian De Ro","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (8 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":"evidence-ap-matt-clancy-clancy-besiroglu-2023-great-inflection","sourceRole":"primary","support":"supports","locator":"Page head: author byline","passage":"Matt Clancy","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-v3-2-d5-cadence-1","sourceRole":"primary","support":"supports","locator":"PDF p. 11 (printed p. 10), §3.1 Scope and Timing, paragraph 'Timing'","passage":"The edition says a Risk Report will be published every three to six months.","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-fsf-v3-0-d7-external-1","sourceRole":"primary","support":"supports","locator":"PDF p. 16 (printed p. 16), §5.2 Disclosures, first paragraph, first sentence","passage":"The edition says that if Google DeepMind assesses a model as having reached a CCL presenting a material, unmitigated risk to public safety overall, it aims to give relevant information to appropriate government authorities where doing so will help the safety of frontier AI.","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 left out the condition that sharing will facilitate the safety of frontier AI. 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 section adds that disclosure to other external organisations may also be considered. Human editorial review is pending."},{"id":"ev-cmp-faif-2026-d1-autonomy-1","sourceRole":"primary","support":"supports","locator":"PDF p. 2, §2.1 Risk Identification, third paragraph, first sentence","passage":"The edition names four primary risk domains, those it considers most likely to lead to significant or severe outcomes in present and future systems: “CBRN Risks”, “Offensive Cybersecurity Risks”, “Loss of Control Risks” and “Harmful Manipulation Risks”.","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":"ev-cmp-faif-2026-d1-autonomy-2","sourceRole":"primary","support":"supports","locator":"PDF p. 1, §1 Introduction, third paragraph, third sentence","passage":"The edition describes loss-of-control risks as those arising when humans can no longer reliably direct, modify or switch off a model.","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":"ev-cmp-ms-fgf-2026-d8-process-1","sourceRole":"primary","support":"supports","locator":"PDF p. 10 (printed p. 9), §5 Governance, 'Updates to this framework', second sentence","passage":"The edition says Microsoft will discuss explicitly, at least every twelve months, how the framework may need improving.","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-ms-fgf-2026-d8-process-2","sourceRole":"primary","support":"supports","locator":"PDF p. 10 (printed p. 9), §5 Governance, 'Updates to this framework', fourth sentence","passage":"The edition says Microsoft's Chief Responsible AI Officer will review any updates to its practices before they are adopted.","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-ms-fgf-2026-d8-process-3","sourceRole":"primary","support":"supports","locator":"PDF p. 10 (printed p. 9), §5 Governance, 'Updates to this framework', fifth sentence","passage":"The edition says every material revision will be published within 30 days of its adoption.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-27","note":"Explicit paraphrase, written on 2026-10-27 by an automated research assistant after re-reading the passage in the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. The comparison quoted the passage; it is given here as a paraphrase to keep the edition within the quotation limit. Human editorial review is pending."},{"id":"evidence-fl-eu-opinion-chapter-coverage-1","sourceRole":"primary","support":"supports","locator":"Paragraph (28) (p. 7)","passage":"the Transparency and Copyright Chapters adequately cover the obligations provided for in Article 53(1)","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (14 words) matched the text retrieved for the candidate pack on 2026-10-27 and again at admission on 2026-10-28, in a fresh retrieval whose SHA-256 equals the pack’s. Read directly from ec.europa.eu. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-eu-opinion-chapter-coverage-2","sourceRole":"primary","support":"supports","locator":"Paragraph (38) (p. 9)","passage":"the Safety and Security Chapter of the Code adequately covers the obligations provided for in Article 55(1)","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (17 words) matched the text retrieved for the candidate pack on 2026-10-27 and again at admission on 2026-10-28, in a fresh retrieval whose SHA-256 equals the pack’s. Read directly from ec.europa.eu. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-eu-opinion-chapter-coverage-3","sourceRole":"primary","support":"supports","locator":"Paragraph (28), second half (p. 7)","passage":"Paragraph (28) adds that the Safety and Security chapter adequately covers the transparency obligations under Annex XI, Section 2, and that the assessment of the Copyright chapter is limited to its adequacy as a means of demonstrating compliance with Article 53(1)(c).","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 ec.europa.eu; it is not a quotation and was not machine-matched. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-orb-oc3-lumen-2024-1","sourceRole":"primary","support":"supports","locator":"PDF p. 2, \"Why Data Centers in Space?\"","passage":"they can be scaled almost indefinitely without the physical or permitting constraints faced on Earth","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (15 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 85) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-orb-oc11-aili-2025-1","sourceRole":"primary","support":"supports","locator":"PDF p. 7, “End of life-cycle”","passage":"Retired hardware ends up as space junk and eventually could burn up in the atmosphere, releasing a small amount of scope 3 emissions and other harmful chemicals","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Added at the admission of the compute, power and orbit pack (research/compute-power-orbit). Located by the cross-family check (earlier-text search, OC11) and read at admission. The quotation (27 words) matched a direct retrieval of 2026-10-30 (SHA-256 399d7be4246bb2b7b1746da24035443ecc54bc23078d75fce02afe4b5fe3ba65). Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-orb-oc11-aili-2025-2","sourceRole":"primary","support":"supports","locator":"PDF p. 7, “End of life-cycle”","passage":"This is a disadvantage compared with terrestrial data centres, where retired components can be readily reused or recycled","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Added at the admission of the compute, power and orbit pack (research/compute-power-orbit). The comparison the passage draws. The quotation (18 words) matched a direct retrieval of 2026-10-30 (SHA-256 399d7be4246bb2b7b1746da24035443ecc54bc23078d75fce02afe4b5fe3ba65). Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-gov-bwc-uk-wp4-alphafold-guardrails-1","sourceRole":"primary","support":"supports","locator":"Timeline of scientific and technological advances, entry for 2021, PDF p. 3","passage":"highlighting the importance of stronger guard rails and AI governance","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (10 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. Explicit paraphrase of the rest of the entry: in 2021 DeepMind published the AlphaFold database, an artificial intelligence programme for protein folding prediction, which accelerated scientific research in fields such as drug discovery but, the paper says, also showed a potential for misuse. No human editorial review.","retrievalRoute":"direct"}],"at":{"entities":[1,10,34,44,47,51,52,55,57,70,156,202,247,299,309,314,341,345,350,356,362,374,411,416,417,503,507,514,517,523,532,533,562,567,585,668,772,865,1008,1091,1105,1117,1125,1205],"claims":[84,150,332,409,567,892,908,992,998,999,1004,1035,1073,1155,1239,1270,1336,1358,1482,1804,1995,2206,2560,2621,2683,2728,2889,3084,3117,3347],"sources":[30,38,41,46,47,50,52,74,75,148,184,246,302,312,318,319,378,383,384,472,474,483,484,515,520,679,680,920,945,1040],"evidence":[99,100,169,354,441,634,1109,1125,1218,1224,1225,1230,1262,1306,1307,1400,1401,1402,1403,1550,1588,1589,1667,1696,1849,2321,2578,2579,2853,3215,3289,3367,3368,3426,3427,3428,3717,3718,3719,4051,4111,4112,4596]},"version":"1lg47t7"}