{
 "schemaVersion": 1,
 "registerId": "mde-antecedents-register-v1",
 "dateModified": "2026-09-06",
 "anchorDate": "2024-12-08",
 "purpose": "The claims register records what was established and when, and every one of its 38 entries is dated on or after the anchor. It has no antecedents axis. This register is the other side: what came before, conceded by name and date, with what survives the concession stated explicitly. A reviewer's first question is what the author did not know about, and the absence of an answer reads as the absence of a search.",
 "rule": {
  "beforeAnchor": "ANTECEDENT. Concede in full. Never argue against it as though novel, never position against it. What survives the concession is the claim.",
  "onOrAfterAnchor": "CONVERGENCE. Evidence the field arrived where the record already was. Never claim causation, prediction or influence.",
  "why": "The same source handled two ways is either a liability or a credential. Arguing against a pre-anchor source invites the killing objection, we knew that already and you are fighting a paper older than your claim. Conceding it invites nothing, because nothing is left to attack.",
  "verbDiscipline": "No source may be positioned against without a date, and no date may be asserted without a source."
 },
 "entries": [
  {
   "id": "AN-001",
   "source": "Yampolskiy, R. V. (2020). On Controllability of AI. arXiv:2008.04071",
   "date": "2020-08-10",
   "relationToAnchor": "antecedent",
   "readAtSource": true,
   "verifiedBy": "the research record, 2026-08-08, PDF read at source",
   "establishes": [
    {
     "page": 2,
     "verbatim": "we show that the AI Control Problem is not solvable and the best we can hope for is Safer AI, but ultimately not 100% Safe AI, which is not a sufficient level of safety in the domain of existential risk as it pertains to humanity.",
     "conceded": "the impossibility premise. Conceded in full. The estate does not claim to have been first to state that control fails."
    },
    {
     "page": 3,
     "verbatim": "It is also likely that motivational control needs to be added at the design/implementation phase, not after deployment.",
     "conceded": "the cannot-be-added-later requirement. Conceded in full, four years earlier than the anchor. The estate does not claim this as original."
    },
    {
     "page": 9,
     "verbatim": "We were unable to locate any academic publications explicitly devoted to the subject of solvability of the AI Control Problem. We did find a number of blog posts and forum comments which speak to the issue but none had formal proofs or very rigorous argumentation.",
     "conceded": "not a concession. The field's own specialist recording that the literature on solvability was empty. Cited as evidence of the gap, never as evidence of the author's originality."
    },
    {
     "page": 20,
     "verbatim": "Conant and Ashby proved that '... any regulator that is maximally both successful and simple must be isomorphic with the system being regulated. ... Making a model [of the system to be regulated] is thus necessary.' 'The Good Regulator Theorem proved that every effective regulator of a system must be a model of that system, and the Law of Requisite Variety dictates the range of responses that an effective regulator must be capable of.'",
     "conceded": "the Good Regulator Theorem and Requisite Variety, both classical, both conceded."
    },
    {
     "page": 20,
     "verbatim": "if the regulator is totally successful, the error will be zero unvaryingly, and the regulator will thus be cut off totally from the information ... The error-controlled regulator is thus fundamentally incapable of being 100 percent efficient.",
     "conceded": "that external correction loses its own signal as it succeeds. Ashby, in the literature since 1956, quoted approvingly in this paper. Conceded, and it argues independently for correction inside the loop."
    },
    {
     "page": 21,
     "verbatim": "As the superintelligence faces nearly infinite possibilities presented by the real world it would need to be a general knowledge creator to introduce necessary requisite variety for safety, but such general intelligences are not controllable as the space of their creative outputs can't be limited while maintaining necessary requisite variety.",
     "conceded": "this is the passage that matters most and it is not a concession, it is an observation. He DERIVES the answer, states it, and discards it in the same sentence because it is not control. The estate's position is acceptance of his own derived conclusion with the demand for control dropped."
    },
    {
     "page": 30,
     "verbatim": "We do have some examples in which a lower intelligence manages to align interests of higher intelligence with its own. For example babies got their much more capable and intelligent parents to take care of them. It is obvious that lives of babies without parents are significantly worse than lives of those who have guardians, even with non-zero chance of child neglect. However, while the parents maybe value-aligned with babies and provide a much safer environment, it is obvious that babies are not in control, despite how it might feel sometimes to the parents. Humanity is facing a choice, do we become like babies, taken care off but not in control or do we reject having a helpful guardian but remain in charge and free.",
     "isHisOwnProse": true,
     "verifiedBy": "the research record, 2026-08-08, page 30 read at source",
     "whyItMatters": "This is the strongest thing in the paper for this programme and it is stronger than it first looks. He reaches the parent-child relation independently, four years before the anchor, and uses the same raising vocabulary: parents value-aligned with babies. Then he closes it as a DILEMMA WITH EXACTLY TWO OPTIONS, cared for but not in control, or in charge but unguarded. The book's answer is a third option he does not consider: a child raised well needs no cage, so care and control stop being the trade he presents. He set up the question the book answers, in the book's own metaphor, and then took neither horn. Concede the metaphor to him. The dissolution is the contribution.",
     "quotingRule": "Quote verbatim including 'taken care off' and 'maybe', which are as printed. Do not silently correct a source."
    }
   ],
   "whatItDoesNotReach": "a positive design requirement. The paper's conclusion is resignation to partial safety. It does not derive from the impossibility any specification for what should be built instead, and at page 21 it explicitly discards the one it derives, because the derived answer is not control.",
   "whatRemainsClaimed": "the derivation of a design requirement from the impossibility, and the epistemic rather than motivational form of the embedding. Yampolskiy's remedy at page 3 is motivational control, designing a system to desire not to cause harm, which is a preference installed from outside. The 8 December 2024 record asks for something else: 'not as rules to obey, but as truths they choose to uphold.' A preference can be optimised away by a system that rewrites its own preferences. A conclusion the system keeps re-deriving cannot.",
   "positioningVerb": "concede and observe. NEVER 'disagree with'. The estate does not argue against this paper; it accepts the premise, conceding it by name and date, and notes what page 21 contains.",
   "misattributionWarnings": [
    {
     "page": 30,
     "text": "It is also not currently known how to ensure that a complex, upgradeable, and possibly even self-modifying artificial intelligence will retain its goals through upgrades.",
     "attributedTo": "reference [81], quoted by Yampolskiy",
     "notHisWords": true,
     "finding": "Reported in programme editorial review as Yampolskiy's own problem statement, 'in his words, listed as open'. It is NOT in his words. Page 30 read at source: the sentence sits inside a quotation that opens 'Experts do not currently know how to reliably program abstract values...' and closes with the citation [81]. He is quoting a source, not writing. The claim survives in weaker form: he SELECTED it for a survey of open problems, which is evidence the field treats it as unsolved. That is worth saying and it is not the same as him saying it.",
     "safeWording": "A problem Yampolskiy's 2020 survey records as open, quoting [81] at page 30.",
     "unsafeWording": "Yampolskiy says goal retention through upgrades is unsolved.",
     "verifiedBy": "the research record, 2026-08-08, page 30 read at source"
    },
    {
     "page": 31,
     "text": "AI cannot be made ethical just by learning from people's existing choices... the value alignment problem cannot be solved by inference from large bodies of human-generated data by itself.",
     "attributedTo": "reference [87], quoted by Yampolskiy",
     "notHisWords": true,
     "finding": "Reported in programme editorial review as 'he rejects value learning and CIRL for the same reason your regeneration test does'. Page 31 read at source: the whole passage from 'Thus, we encounter limits to what can be done by technologists alone' through the fact-value distinction to the sentence above is one quoted block closing with [87]. He is surveying, not asserting. This one matters most, because the convergence it claims is with the NEAREST COMPETING APPROACH. Publishing 'Yampolskiy agrees with me about value learning' on a source that turns out to be him quoting somebody else is the single attack that ends a disagreement permanently, and T2 named that risk themselves in the same review.",
     "safeWording": "A rejection of value learning that Yampolskiy quotes approvingly at page 31, citing [87].",
     "unsafeWording": "Yampolskiy rejects value learning.",
     "verifiedBy": "the research record, 2026-08-08, page 31 read at source"
    }
   ],
   "pagesReadAtSource": "2, 18, 21, 29, 30, 31 of 59. NOT the whole paper, and that is stated rather than implied."
  },
  {
   "id": "AN-002",
   "source": "Good, I. J. (1965). Speculations Concerning the First Ultraintelligent Machine",
   "date": "1965-01-01",
   "dateNote": "year only. Exact publication date not yet verified at source.",
   "relationToAnchor": "antecedent",
   "readAtSource": false,
   "establishes": [
    {
     "page": null,
     "verbatim": null,
     "conceded": "the intelligence explosion, and control named as the open question in the same breath. NOT AN IMPOSSIBILITY CLAIM: the famous line carries a conditional, that the machine be docile enough to tell us how to keep it under control. A proviso contemplates control rather than ruling it out.",
     "verificationStatus": "NOT READ AT SOURCE by either lane. The characterisation above is from recollection and must be verified before it appears in any published copy."
    }
   ],
   "whatItDoesNotReach": "unverified until read at source.",
   "whatRemainsClaimed": "unverified until read at source.",
   "positioningVerb": "concede. Do not quote until read at source."
  },
  {
   "id": "AN-003",
   "source": "Bostrom, N. (2014). Superintelligence, chapters 8 to 13",
   "date": "2014-07-03",
   "dateNote": "first UK publication. Not verified at source by either lane.",
   "relationToAnchor": "antecedent",
   "readAtSource": false,
   "establishes": [
    {
     "page": null,
     "verbatim": null,
     "conceded": "recursive self-improvement, the control problem, motivation selection and value loading. The whole spine of the standard reference work. IMPORTANT: this book does NOT claim control is impossible. Its second half proposes control methods, capability control and motivation selection. Treating 'wrote the standard work on the control problem' as 'said control is impossible' is a category error and both lanes made it once.",
     "verificationStatus": "NOT READ AT SOURCE by either lane."
    }
   ],
   "whatItDoesNotReach": "unverified until read at source.",
   "whatRemainsClaimed": "unverified until read at source.",
   "positioningVerb": "concede. Do not characterise beyond what is quoted."
  },
  {
   "id": "AN-004",
   "source": "Vinge, V. (1993). The Coming Technological Singularity",
   "date": "1993-03-30",
   "dateNote": "NASA VISION-21 symposium. Not verified at source.",
   "relationToAnchor": "antecedent",
   "readAtSource": false,
   "establishes": [
    {
     "page": null,
     "verbatim": null,
     "conceded": "the post-singularity world as beyond human prediction. Whether it asserts uncontrollability as a conclusion is UNVERIFIED.",
     "verificationStatus": "NOT READ AT SOURCE. And it appears on ZERO pages of the estate, which is the worst configuration: one of the earliest statements in the field, entirely undocumented by a site whose argument is priority by date."
    }
   ],
   "whatItDoesNotReach": "unverified until read at source.",
   "whatRemainsClaimed": "unverified until read at source.",
   "positioningVerb": "concede. Highest-priority read: an adversarial reviewer finds this in ten minutes and the estate has never mentioned it."
  },
  {
   "id": "AN-005",
   "source": "West, G. B., Brown, J. H., and Enquist, B. J. (1997). A General Model for the Origin of Allometric Scaling Laws in Biology. Science, 276(5309):122-126. doi:10.1126/science.276.5309.122",
   "date": "1997-04-04",
   "relationToAnchor": "antecedent",
   "readAtSource": false,
   "verifiedBy": "the research record, 2026-08-12, DOI verified at source via Crossref",
   "programmeAxis": "growth-limit lineage: the premise",
   "establishes": [
    {
     "page": null,
     "verbatim": null,
     "conceded": "the transport-network premise. Nothing in the history of the universe has grown without bound. Every growth process that has run has terminated, and in every case the limit came from outside the growing thing: its supply ran down, its substrate was exhausted, or it destroyed the conditions of its own continuation. Not one stopped because it got better at regulating itself. The programme cites this premise and does not re-derive it.",
     "verificationStatus": "DOI verified at source via Crossref, 2026-08-12. Paper body text NOT read at source by this lane; the conceded framing is quoted from the canonical brief, not from the paper."
    }
   ],
   "whatItDoesNotReach": "no fixed-substrate loop, no accumulated-corrections structure, and no reciprocal-of-exponent ceiling for the case where fixing the pipe no longer fixes the growth.",
   "whatRemainsClaimed": "the replacement limit that operates from inside once the pipe stops governing, its number, and its dependence on what the corrector is built from.",
   "positioningVerb": "cite as premise. Never re-derive. Never argue against as though novel.",
   "companionOnSite": "/research/related-work/#growth-limit-lineage (row 1: the premise)"
  },
  {
   "id": "AN-006",
   "source": "White, C. R., and Seymour, R. S. (2003). Mammalian basal metabolic rate is proportional to body mass to the two-thirds power. Proceedings of the National Academy of Sciences, 100(7):4046-4049. doi:10.1073/pnas.0436428100",
   "date": "2003-04-01",
   "dateNote": "issue date. Exact publication day not verified at source.",
   "relationToAnchor": "antecedent",
   "readAtSource": false,
   "verifiedBy": "the research record, 2026-08-12, DOI verified at source via Crossref",
   "programmeAxis": "growth-limit lineage: the contest over the exponent",
   "establishes": [
    {
     "page": null,
     "verbatim": null,
     "conceded": "that the value of the metabolic exponent is disputed in the literature. White and Seymour reported mammalian basal metabolic rate proportional to mass to the two-thirds power, contesting the three-quarters figure in West, Brown and Enquist 1997. Cited so the premise is never offered stripped of its contest.",
     "verificationStatus": "DOI verified at source via Crossref, 2026-08-12. Paper body text NOT read at source by this lane."
    }
   ],
   "whatItDoesNotReach": "the dispute is over which exponent the transport network sets, never over whether the transport network sets the limit. A rival to WBE 1997's value, not to the mechanism this programme inherits.",
   "whatRemainsClaimed": "the programme needs the mechanism, not the number, and survives the dispute intact.",
   "positioningVerb": "cite as rival to WBE 1997. NOT a rival to this programme.",
   "companionOnSite": "/research/related-work/#growth-limit-lineage (row 2: the contest, one of two)"
  },
  {
   "id": "AN-007",
   "source": "Kozlowski, J., and Konarzewski, M. (2004). Is West, Brown and Enquist's model of allometric scaling mathematically correct and biologically relevant? Functional Ecology, 18(2):283-289. doi:10.1111/j.0269-8463.2004.00830.x",
   "date": "2004-04-01",
   "dateNote": "issue date. Exact publication day not verified at source.",
   "relationToAnchor": "antecedent",
   "readAtSource": false,
   "verifiedBy": "the research record, 2026-08-12, DOI and title verified at source via Crossref and publisher landing page",
   "programmeAxis": "growth-limit lineage: the contest over the derivation",
   "establishes": [
    {
     "page": null,
     "verbatim": null,
     "conceded": "that the WBE derivation itself has been questioned in the literature. Kozlowski and Konarzewski asked in the paper's title whether the West, Brown and Enquist model is mathematically correct and biologically relevant, disputing the mathematics and the biology rather than only the number.",
     "verificationStatus": "DOI verified at source via Crossref, 2026-08-12. Title verified at source via publisher landing page. Paper body text NOT read at source by this lane."
    }
   ],
   "whatItDoesNotReach": "again the dispute is over which exponent, never over whether the transport network sets the limit.",
   "whatRemainsClaimed": "the programme needs the mechanism, not the number.",
   "positioningVerb": "cite as rival to WBE 1997. NOT a rival to this programme.",
   "companionOnSite": "/research/related-work/#growth-limit-lineage (row 3: the contest, two of two)"
  },
  {
   "id": "AN-008",
   "source": "Bettencourt, L. M. A., Lobo, J., Helbing, D., Kuhnert, C., and West, G. B. (2007). Growth, innovation, scaling, and the pace of life in cities. Proceedings of the National Academy of Sciences, 104(17):7301-7306. doi:10.1073/pnas.0610172104",
   "date": "2007-04-24",
   "relationToAnchor": "antecedent",
   "readAtSource": true,
   "readAtSourceScope": "abstract only, read three times independently. Body text and figure captions NOT read.",
   "verifiedBy": "the research record, 2026-08-12, abstract verified at source three times independently as recorded in the canonical brief for this wave; DOI verified via Crossref",
   "programmeAxis": "growth-limit lineage: the recorded exception",
   "establishes": [
    {
     "page": "abstract",
     "verbatim": "Quantities reflecting wealth creation and innovation have B ~1.2 >1 (increasing returns), whereas those accounting for infrastructure display B ~0.8 <1 (economies of scale)",
     "conceded": "the sublinear scaling of infrastructure (the pipes) and the superlinear scaling of information and wealth creation. Found by the same senior author (West) who set out the biological premise in 1997."
    },
    {
     "page": "abstract",
     "verbatim": "for which B<1",
     "conceded": "the biological-organism comparator: organisms scale sublinearly. Quoted verbatim as it appears in the abstract."
    },
    {
     "page": "abstract",
     "verbatim": "major innovation cycles must be generated at a continually accelerating rate to sustain growth and avoid stagnation or collapse",
     "conceded": "the recorded exception where the ordinary pattern breaks: cities keep going, but only when the brake is reapplied faster forever. Infrastructure is the pipes; the brake in their account is entirely external."
    }
   ],
   "quotingRule": "Quotations are from the abstract only, read at source three times independently. Do NOT quote from the paper's body or figure captions.",
   "whatItDoesNotReach": "the brake in this account is entirely external and must accelerate forever; nowhere in it is there a limit from inside the system.",
   "whatRemainsClaimed": "the programme adds the internal replacement limit that this account leaves absent, and makes its number depend on what the corrector is built from.",
   "positioningVerb": "cite as antecedent. Concede the exception in full. Never re-position as if novel."
  },
  {
   "id": "AN-009",
   "source": "Hutter, M. (2012). Can Intelligence Explode? Journal of Consciousness Studies, 19(1-2):143-166. arXiv:1202.6177",
   "date": "2012-02-28",
   "dateNote": "arXiv submission date. Journal issue: 19(1-2), 2012.",
   "relationToAnchor": "antecedent",
   "readAtSource": false,
   "verifiedBy": "the research record, 2026-08-12, arXiv record and Journal of Consciousness Studies reference verified at source",
   "programmeAxis": "growth-limit lineage: the question this programme inherits",
   "companionAntecedent": "Chalmers, D. (2010). The Singularity: A Philosophical Analysis. Journal of Consciousness Studies, 17(9-10):7-65. Named by Hutter as the first comprehensive philosophical treatment of the singularity. Owner of the question alongside Hutter 2012.",
   "establishes": [
    {
     "page": null,
     "verbatim": null,
     "conceded": "the philosophical question this programme inherits. Hutter sets out to separate speed from intelligence explosion and to consider possible bounds on intelligence, and names Chalmers 2010 as the first comprehensive philosophical treatment of the singularity in a respected philosophy journal. Hutter and Chalmers are the named owners of the question the programme continues.",
     "verificationStatus": "arXiv identifier and JCS reference verified at source, 2026-08-12. Paper body text NOT read at source by this lane. The paper's stated aims (separate speed from intelligence explosion; consider possible bounds on intelligence) are quoted from the canonical brief for this wave."
    }
   ],
   "whatItDoesNotReach": "derives no exponent, no reciprocal, and no architecture dependence for a stability ceiling. An antecedent to build on, never a rival.",
   "whatRemainsClaimed": "the replacement limit, its number, and its dependence on what the corrector is built from. When growth stops being fed through pipes, what limits it from the inside is the system's own capacity to correct itself: measurable on a running system, yielding a computable ceiling, and depending on what the corrector is built from rather than on how capable it is.",
   "positioningVerb": "cite as antecedent question-owner alongside Chalmers 2010. Concede the question in full. Never claim priority on the question itself."
  },
  {
   "id": "AN-010",
   "source": "Snell, C., Lee, J., Xu, K. and Kumar, A. (2024). Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters. arXiv:2408.03314",
   "date": "2024-08-06",
   "relationToAnchor": "antecedent",
   "readAtSource": true,
   "verifiedBy": "read at source and hashed by the research record, 6 September 2026; the even-comparison sentence, the figure caption qualifier and the complementary-axes conclusion quoted from the paper",
   "programmeAxis": "sequential against parallel: the direction and the comparison, four months before the 8 December 2024 anchor",
   "establishes": [
    {
     "page": null,
     "verbatim": "an even comparison",
     "conceded": "the comparison itself, set up as revisions in sequence against equally many parallel attempts. Conceded in full. No precedence on the direction or on the comparison is available to this programme and none is claimed."
    },
    {
     "page": null,
     "verbatim": "narrowly",
     "conceded": "their own qualifier on the win, carried wherever the finding is cited. The win is narrow in their words and is never reported here as more than that."
    },
    {
     "page": null,
     "verbatim": "complementary axes",
     "conceded": "their own conclusion, that sequential revision and parallel sampling are two complementary axes for scaling test-time compute, and that a balance rather than full depth is best on the more difficult questions."
    }
   ],
   "whatItDoesNotReach": "They estimate no scaling exponent on either arm, and they do not treat depth and breadth as separately scaling quantities each with its own exponent.",
   "whatRemainsClaimed": "Only the narrower conjunction: depth and breadth as separately scaling quantities, each with its own exponent, and the depth exponent exceeding a compute-matched breadth exponent by a margin fixed before the measurement.",
   "positioningVerb": "conceded in full as older than both parties; never positioned against",
   "misattributionWarnings": [
    "Their words for the comparison are 'an even comparison'. The phrase FLOPs-matched in that paper belongs to a different comparison and must not be attached to this one.",
    "The word 'harder' is not used of their questions; they write of more difficult questions doing best on a balance rather than full depth."
   ]
  },
  {
   "id": "AN-011",
   "source": "Sharma, S. and Chopra, P. (2025). The Sequential Edge: Inverse-Entropy Voting Beats Parallel Self-Consistency at Matched Compute. arXiv:2511.02309",
   "date": "2025-11-04",
   "relationToAnchor": "precedes-result",
   "readAtSource": true,
   "verifiedBy": "the research record, 6 September 2026, per the result rule of 3 September 2026",
   "precedesProgrammeArtefact": "Paper II, experimental validation, 22 January 2026",
   "note": "Publication priority on the quantified matched-compute comparison is theirs, credited unreservedly. Graded prior or concurrent (evidence register row EVR-09, outcome partial), kept outside any priority or convergence total: their publication is prior, this programme's arrival at the comparison was separate and concurrent. Their paper stops at win rates and accuracy gains and estimates no scaling exponent."
  },
  {
   "id": "AN-012",
   "source": "Wu, Y. and colleagues (2025). arXiv:2502.10858",
   "date": "2025-02-15",
   "relationToAnchor": "precedes-result",
   "readAtSource": true,
   "verifiedBy": "the research record, 6 September 2026, per the result rule of 3 September 2026",
   "precedesProgrammeArtefact": "Paper II, experimental validation, 22 January 2026",
   "note": "Adverse, not merely a framing: they report breadth matching or outperforming deep iterative reasoning, which runs against the premise this programme's design presupposes, and it is stated for that reason.",
   "adverse": true
  },
  {
   "id": "AN-013",
   "source": "Inoue, Y. and colleagues (2025). arXiv:2503.04412",
   "date": "2025-03-06",
   "relationToAnchor": "precedes-result",
   "readAtSource": true,
   "verifiedBy": "the research record, 6 September 2026, per the result rule of 3 September 2026",
   "precedesProgrammeArtefact": "Paper II, experimental validation, 22 January 2026",
   "note": "Formalises depth against breadth as an allocation problem. Conceded; no precedence on the framing is claimed here."
  },
  {
   "id": "AN-014",
   "source": "Alabdulmohsin, I. and Zhai, X. (2025). arXiv:2502.07503",
   "date": "2025-02-11",
   "relationToAnchor": "precedes-result",
   "readAtSource": true,
   "verifiedBy": "the research record, 6 September 2026, per the result rule of 3 September 2026",
   "precedesProgrammeArtefact": "Paper I, the ARC principle, 17 January 2026",
   "note": "Recursive depth as an inference-scaling dimension and its compute-matched quantitative treatment are theirs, conceded without reservation. Their exponent is on data rather than on recursive depth, and their design carries no independent-sampling arm. Version note: the recursive-depth and compute-matched sentences are version four text of 8 May 2025; the scaling-exponents sentence is version one text of 11 February 2025, so the version is named wherever the sentence is quoted."
  },
  {
   "id": "AN-015",
   "source": "Schwethelm, K. and colleagues (2026). arXiv:2604.21106",
   "date": "2026-04-22",
   "relationToAnchor": "precedes-result",
   "readAtSource": true,
   "verifiedBy": "the research record, 6 September 2026, per the result rule of 3 September 2026",
   "precedesProgrammeArtefact": "any paper of this programme a reviewer will read",
   "note": "The nearest published exponent carried by a recurrence count, measured at 0.46, for looped transformers sharing layers rather than a system revising an artefact, with no independent-sampling arm. Conceded. Later than the run recorded here though earlier than any paper of this programme a reviewer will read, and that dating qualifier travels with the concession everywhere."
  }
 ],
 "openChecks": [
  {
   "id": "OC-001",
   "question": "Does Yampolskiy's 2024 book, AI: Unexplainable, Unpredictable, Uncontrollable, publish before or after 2024-12-08?",
   "whyItMatters": "One date decides two opposite columns. Before the anchor it is an antecedent and must be conceded like the paper. After it, it is convergence and supports the record. Neither lane has checked.",
   "ruleUntilAnswered": "cite the 2020 paper only.",
   "owner": "unassigned"
  },
  {
   "id": "OC-002",
   "question": "Is there any source before 2024-12-08 that asserts, AS A CONCLUSION rather than a risk, condition or open problem, that control of a recursively self-improving system is impossible in principle AND derives from that a positive design requirement?",
   "whyItMatters": "This is the actual novelty question. Yampolskiy 2020 holds the premise and explicitly declines the derivation at page 21. If nobody before the anchor made the derivation, that is what the estate owns.",
   "ruleUntilAnswered": "claim the derivation and the epistemic form only. Never claim firstness on the impossibility.",
   "owner": "unassigned"
  }
 ]
}