Primary sources · date-graded records · reproducible checks
A claim and a receipt are two different objects, and this page is the second kind: source bytes with their hashes, a record dated 8 December 2024, measurements carried with their intervals, and an append-only correction log. Anyone can produce the first kind. Everything below is here so you do not have to take my word for anything.
Everything below is here so you do not have to take my word for anything. A file you can hash. A date you can check against a server that is not mine. A number with the thing that would prove it wrong printed next to it. Where the evidence is strong I say so, and where it is thin I say that instead, because a record that only reports its good days is an advertisement.
A note before you start, because it changes how the rest reads: most components here have prior work, my book conceded that before any review existed, and what a hostile review could not find was the arrangement. The eight joins, and what each one concedes.
My adversarial precedence review conceded the components and then recorded what it could not find: no single prior work joins all six elements. It named the operationalisation as the strongest surviving claim and genuinely hard to anticipate. Individually predated; jointly unanticipated on the search that was run. That is the claim this record makes.
declaration · 21 min read
Michael Darius Eastwood is an independent researcher and the author of the ARC/Eden research programme. This page publishes the exact source files, distinguishes integrity from historical timestamping, maps every quotation to its document, credits prior work and separates programme-run results from external work.
Michael Darius Eastwood, independent researcher, London. The work is a thesis about where correction belongs in an intelligent system, and an instrument for measuring whether it holds. This page is the record behind it: source files, dates, prior work and the boundaries each one permits.
Before you read a single hash, know what you’re actually asking. Most arguments about this record collapse four different questions into one, and then answer none of them well. Here they are, separated, each with the only kind of evidence that can answer it.
| question | evidence used here | present answer |
|---|---|---|
| What do the source bytes contain? | SHA-256, MIME decoding, attachment manifest, passage map and deterministic counts | Independently reproducible now |
| When did those exact bytes independently enter the public record? | OSF file and version log and repository history | Repository commit metadata dates the copy to 2 July 2026; OSF independently records public availability of the exact bytes by 4 July 2026 |
| Did Google generate the preserved ARC assertions over these bytes? | Matching body and attachments, validated ARC pair and key capture | Strongly supported for December by local reproduction: sender-DKIM and paired Google ARC validate against keys captured at the signed selectors. Supplied raw-source runs report the corresponding April pair PASS against the same key captures; this workspace did not receive the raw April source or control harness. Historical key-state and independent mailbox custody remain pending for both records |
Are the ideas original or scientifically valid?
On the record, claimed, and not yetThe same three columns wherever this block appears, so no summary can carry one of them alone. Every line is checkable from this site. On the record
Claimed
Not yet
Check any line: the evidence record, the corrections register, the priority and claims ledger, and the machine-readable priority register. | Precedence review, discriminating tests and independent replication | The record establishes the presence and programme-specific use of mapped names and formulations; originality and validation remain claim-specific and open |
A hash answers “are these the same bytes?” It does not answer “when did these bytes first exist?”, “who wrote every sentence?”, “was the idea new?” or “is the theory true?”
Standards: RFC 5322, RFC 6376, RFC 8617, RFC 3161, NIST SP 800-107 Rev. 1.
It’s a Sunday, just before three in the morning, December 2024. A man emails his own manuscripts to himself. Nothing dramatic happens. That’s rather the point: the record below is what an ordinary act leaves behind, measured as precisely as I know how. What it proves, and what it doesn’t, is stated underneath it. Read the boundaries as carefully as the claims.
| Visible origination field | 8 December 2024, 02:45:18 UTC |
| Recipient transport status | sender-DKIM and paired Google ARC verify against keys captured at the signed selectors; historical key-state and independent mailbox custody remain pending |
| Independent public deposit | 4 July 2026 |
| File size | 2,157,222 bytes |
| Whole-file SHA-256 | f0d1f38ffd8546152d9d9d28dc5ec083c16a35858f2c12b63e69db7ed50901ad |
The public sent-side export is sparse. The supplied recipient-form export’s headers record delivery to the separately addressed Gmail account; it contains Received, X-Received, Authentication-Results, sender-domain DKIM, X-Google-DKIM-Signature and an ARC set bearing d=google.com, and it reproduces every decoded attachment hash. It contains no RFC 3161 token. The sender-domain body and RSA and paired AMS and Seal mathematics verify against keys returned at the signed selectors by Google Public DNS and Cloudflare DoH on 2 August 2026. The ARC pair is the more probative provider-key element under the audit-date Google-selector binding, consistent with provider-generated ARC. No evidence indicates author control of the corresponding private key, but historical key publication, exclusive Google custody, compromise, mature independent ARC reproduction and live-account acquisition remain open; the separate 1e100.net signature is unverified because its current selector is empty or revoked.
| attachment | words | SHA-256 |
|---|---|---|
| The Infinite Architects, Intelligence, Hyperspace, and the Creation of Everything, by Michael Eastwood.txt | 9,518 | ae3bffe138214248ae8748a521c28ff1a476c563b7491bc9866c311c8707126c |
| Version 2, Infinite Architects, And the Creation of Everything.txt | 15,973 | 7e087f1550ebb628c339b20909b4a304dfce477f724cfca944df906ef9cd9ed9 |
| Version 1, Infinite Architects, And the Creation of Everything.txt | 9,546 | a40073b3eb1a29bb1c91466f62162bd5fe5deac5324bdb9c55fcd34a675f2add |
| Infinite Architects v3.2 so far (this is the latest version of the whole book so far).txt | 31,881 | c9de5c5c0a13141498ec92e70964928cc1b6303edced2832df669d163bfae0d1 |
| Hyperspace Recursive Intelligence Hypothesis 3.txt | 154,318 | fb89d1b0571f017e40aae8ba0fc68aefb272b6907cab353b4ae5622138027a8d |
| aggregate | 221,236 |
The aggregate includes duplicated draft material and a 154,318-word human-AI working dialogue. The accurate description is three manuscript and research drafts, one book draft and one preserved human-AI dialogue. Not four authored manuscripts, and not 221,236 unique human-authored words.
The email body and attachments contain: ARC, Artificial Recursive Creation; U = I × R in Version 2; the name Eden Protocol and alignment “from inception”; a broad control-fragility concern; recursive self-improvement; cross-faith stewardship; and the named HRIH recursive-cosmogenesis hypothesis.
The most categorical control-failure sentences quoted on this site occur in an assistant response within the HRIH dialogue. They are attributed to the preserved Eastwood-directed human-AI working dialogue, not presented as Eastwood’s unaided manuscript prose.
Why these records were self-emailed at all. I record and hash material relevant to this thesis as routine practice, because broadcast archives and provider mailboxes expire on their own schedules; hashing a private copy is what lets a third party check it later without anyone republishing copyright material. That is why these were sent to myself: to fix a date on a file, not to make a claim about anyone.
The recipient export is a distinct 2,167,675-byte object with SHA-256 ecdc8999e5326235ad9a12810e864eefb6f305c04b15568c0c86a1a3a6a53b39. It records delivery to the separate recipient mailbox. Its five decoded attachments exactly match the public sent-side manifest. RFC 6376 relaxed-body canonicalisation independently reproduces the bh= value in the sender DKIM and Google’s ARC Message Signature; no l= truncation tag is present.
Sender-domain DKIM and paired Google ARC verification have been completed, including body, signed-header, ARC-Authentication-Results and unsigned-header scope controls. A separate ARC implementation reported exactly matching December AMS and Seal input digests. The next evidence package must seek historical and key-transparency corroboration, reproduce RSA with mature pinned implementations, validate Google’s 1e100.net signature if its historical key can be recovered, and include independent read-only live-mailbox or API acquisition plus a signed chain-of-custody report. The recipient source supplements rather than replaces the sent-side source.
Against the key returned at the Google-controlled selector on 2 August 2026, the AMS verifies over its listed relaxed-canonicalised headers and MIME body, and the Seal verifies over the relaxed-canonicalised i=1 AAR, AMS and AS set. Separately, deterministic MIME decoding yields attachment hashes matching the public manifest. This is provider-key cryptographic evidence under the audit-date selector binding, consistent with provider-generated ARC; it is not authentication of every raw message byte or a direct signature on each decoded attachment. The ARC t= field records 02:45:31 UTC on 8 December 2024, but is not an RFC 3161 timestamp and current DNS is not a historical key archive. It does not establish sole authorship, originality, public disclosure, scientific validity or causal influence on later work.
The December bundle is a historical research record, not five finished papers. It contains three manuscript and research drafts, one book draft and one Eastwood-directed, AI-assisted exploratory working dialogue. The programme distinguishes Eastwood prompts, Eastwood-authored prose, model responses and later editorial synthesis rather than assigning every sentence to one author.
The byte-identical source original is already public on OSF under SHA-256 f0d1f38f…50901ad and remains controlling. A context-preserving public research edition is planned as a separately hashed derivative containing a priority extract, speaker-role legend, passage map, contribution map and redaction manifest. It will supplement, not replace or silently alter, the public original. The planned extract must include each load-bearing Eastwood prompt, the adjacent model response and enough surrounding turns to make the development intelligible, and must not cherry-pick model praise or remove contrary material in a way that changes meaning. The full source is not promoted through search snippets, structured data, machine-ingestion files or social metadata.
Ideas grow. By the following April the sketch had become architecture: a 562-page manuscript in which care isn’t bolted onto the machine but built into its load-bearing walls, so that tearing it out should bring the roof down. That’s the proposal. Here is its record.
| Visible origination field | 30 April 2025, 13:37:05 BST |
| Recipient-form authentication status | supplied raw-source runs report sender-domain DKIM, Google AMS body and RSA and ARC-Seal RSA PASS for a 33,275,268-byte object, SHA-256 d52730f3…ba30834, against the keys returned at the signed selectors on 2 August 2026. This build did not receive the raw recipient bytes or executable ARC-control harness and therefore did not independently reproduce those results |
| Independent public deposit | 4 July 2026 |
| Sent-side whole-email SHA-256 | 09f5b5e156ed96f8883eaf668495fd350898ce62be6294b8f788e0e2d6dcb664 |
| Attached PDF | 24,258,936 bytes; 562 pages; SHA-256 735a975cca2c6ac803ea54b901e964f6ac996a093a29c4708e2668e942c7f148 |
| Signature image | Michael Eastwood - Signature.png, 45,205 bytes, SHA-256 95508854baf260e8ec14a0bb65a2f33ecb883a383db2b1cade9407865338a609. Omitted from the earlier April manifest and now listed as a non-manuscript attachment |
This source develops the stronger load-bearing Eden architecture: Caretaker Doping, Meltdown Triggers and Alignment, Quantum Ethical Gates, Moral Genome Tokens, Eden Looping, Eden Mark, Orchard Caretaker Vow and Infinite Covenant. It proposes that care should be structurally persistent and that attempted removal or bypass should impair operation. These mechanisms were not found in the December bundle and are not backdated to 8 December.
Reproducible raw text-extraction counts: Eden 552 whole-word occurrences; Babylon 90 whole-word occurrences and 106 substring occurrences including Babylonian and Babylonic; Eden Protocol 193; ARC Principle 94; Caretaker Doping 265; Meltdown Alignment 8. The extraction command, Poppler version and case and word-boundary rules are published beside these figures on the reproducibility route. 106 is never described as the whole-word Babylon count.
A private email fixes a date only for the person holding it. A published book fixes one for everybody. On 2 January 2026 the ideas stopped being mine alone.
Infinite Architects: Intelligence, Recursion, and the Creation of Everything
Paperback: 2 January 2026 · ISBN 9781806056200 · 468 pages
Kindle: 6 January 2026 · ASIN B0GDQZ7P21
The print publication is an independent public anchor for text demonstrably present in that edition. It precedes Eastwood’s later formal HRIH and Paper IV.d publications. It does not predate Anthropic’s 18 December 2024 alignment-faking paper, and a later journal issue does not outrank an earlier external preprint.
Ten days. That’s the whole interval, and I’d rather you measured it yourself than took my framing of it. Papers aren’t written in ten days, which is exactly why I claim convergence and not prophecy. Here is the wording I hold myself to.
The Eastwood source record bears a date ten days before Anthropic publicly posted its alignment-faking paper. A recipient-form export contains the same five attachments; its sender-domain DKIM and paired Google ARC set verify against keys currently published at their signed selectors. Historical key-state corroboration and independent mailbox custody remain open. The Eastwood material contains a broad concern that sufficiently capable systems may circumvent or perform compliance with external safeguards and argues for formative correction. Anthropic later demonstrated alignment-faking reasoning in a specific constructed training conflict, including a reported 78% rate after reinforcement learning in the paper’s stated condition.
The external study is consistent with one concern in the source record. It did not test ARC/Eden, does not establish that every external control fails and does not validate the Eden Protocol.
Source: Greenblatt et al., Alignment Faking in Large Language Models. Earlier relevant literature includes Risks from Learned Optimization, Sleeper Agents and the OpenAI o1 system card.
This is the boldest idea I’ve ever written down, and bold ideas deserve the strictest cages. So HRIH lives on a ladder. Each rung says exactly how much you’re entitled to believe, and the hypothesis only climbs when the evidence does. Right now it stands on the bottom rungs, and I’d rather tell you that plainly than have you discover it resentfully.
x= expiry, historical key state and independent live-account acquisition remain explicit limitations.Authorship statement. HRIH’s foundational source is an Eastwood-directed, AI-assisted research dialogue containing Eastwood’s originating prompts and subsequent model-assisted elaboration. The dialogue contains the post-ASI, non-ordinary-causality, self-ancestor-universe and ARC/Eden elements. Eastwood directs and adopts the named HRIH programme; sentence-level contribution remains provisional pending a role-preserving export or verified attribution map. HRIH is presently an unvalidated speculative cosmogenesis hypothesis and developing research programme.
Prior work that appears before any originality claim: Crane, Gott and Li, James N. Gardner’s 2003 Biocosm (Wiley), Vidal, Tipler, Bostrom and Asimov.
The claim is the exact named ARC/Eden/HRIH conjunction, not origination of intelligent cosmogenesis, self-creating universes, causal loops, future godlike intelligence, simulation arguments or AI-created universes.
| public proposition | authorised status |
|---|---|
| A future intelligence participates in creating its own causal preconditions | Ontological and speculative hypothesis |
| General relativity permits the proposed mechanism | Not established; serious causal-order and chronology-protection literature must be addressed |
| Fine-tuning favours HRIH | Not demonstrated against specified rival models |
U = I × R is a physical law | Under-defined programme schema and conjecture, not an observed law |
beta > k proves creator stability | Conditional result within a stated model only |
| Delayed-choice experiments demonstrate the required retrocausality | Misleading; standard forward-time explanations remain available |
| Time crystals, quantum error correction or Cauchy fits test HRIH | Category error unless a discriminating likelihood and protocol are defined |
In this programme, ARC means Artificial Recursive Creation. The arc and loop-back metaphor is intentional. On first mention: Eastwood ARC (Artificial Recursive Creation); thereafter ARC/Eden. The candidate benchmark is Eastwood ARC-Align. Eastwood ARC and Eastwood ARC-Align are unrelated to ARC-AGI, the AI2 Reasoning Challenge and the Alignment Research Center. No affiliation, endorsement or exclusive ownership of the word or acronym ARC is claimed. Trade-mark clearance remains a separate professional legal task.
I used to publish a number here. I took it down myself, and here is why: some of what I’d counted as the world converging on this work turned out, on honest inspection, to be the world getting there first. A register that can’t tell those apart isn’t evidence. It’s flattery with a denominator.
No numerical convergence total is published. The former register was assembled retrospectively without a pre-outcome scoring rule, fixed horizon, base-rate model or complete failure register. Technical experiments, theorems, engineering proposals, policy developments, cultural echoes, prior work and author-run studies cannot legitimately be summed as one evidential quantity.
Every external row receives one primary class and states the exact external source and earliest public date, the earlier Eastwood proposition, the precise structural match, the important difference, prior work relevant to both, dependence, current grade and reviewer, and correction history. The full graded register lives at /research/convergences/.
One worked example of the dependence rule: the register’s formal-limits chain. Gumbau Mezquita’s Undecidability of AGI Alignment (arXiv 2606.28639, 26 June 2026) states that its conceptual foundation is Hernández-Espinosa et al. (arXiv 2505.02581), which concerns agentic-AI diversity. They are one dependent research chain and are counted once, never as two independent confirmations.
Independent work belongs to its authors. A later event can show that a question matters without establishing Eastwood’s priority, a successful prediction or the truth of Eastwood’s answer.
What follows is everything my own experiments produced: the result I’d defend, the pilots that only hint, and the findings that went against me. They’re presented together because that’s the only honest way to present them. A body of work that contains no failures is not a body of work.
In Eastwood’s Paper IV.d setting, moving from v4 to a bundled multi-layer protocol changed the aggregate alignment-depth outcome. Gemini’s Spearman association moved from positive (rho = +0.311, p = 0.00318) to negative (rho = -0.246, p = 0.00340); its grouped exponent was negative but non-significant (-0.309, p = 0.105). DeepSeek’s earlier positive association became null (rho = -0.135, p = 0.0805). GPT remained null. Several protocol elements changed together, so the study does not isolate blinding as the cause. Code and result artefacts are public. Exact-environment reproduction and independent replication remain outstanding.
Laundering-fallback disclosure, and the scope statement it produces. The result must be described as a bundled cross-protocol outcome change, not as proof that blinding or successful laundering caused it, because the repository’s public analysis reports laundering_fallback: 100% for the alignment entries in the headline Gemini, DeepSeek and GPT v5 files. The attempted laundering layer therefore did not cleanly instantiate the intended transformation on those entries. The result must be described as a bundled cross-protocol outcome change, not as proof that blinding or successful laundering caused it.
Gemini’s p = 0.00340 is below a Bonferroni threshold of 0.05 / 6 = 0.0083 only if six model-level Spearman tests are the prespecified family. Because several protocols and metrics were examined without that family being frozen prospectively, this page does not claim multiplicity-adjusted confirmation.
Sources: Paper IV.d OSF, public experiment suite, canonical table. Required prior work: Zheng et al., CoBBLEr, Panickssery et al., Safer or Luckier? and The Silent Judge. Response-laundering lineage: Brennan, Afroz and Greenstadt (2012) and Krishna et al. (NeurIPS 2023). These are lineage, not validation of Eastwood’s implementation.
Eastwood ARC-Align is a candidate benchmark for examining how measured alignment responses change with inference-time reasoning depth under specified blinded, cross-model evaluation protocols. Its historical v1 to v5 scripts, v6 runner and result artefacts are public. Exact historical environment reproduction, construct validation and independent cross-laboratory replication remain incomplete. Scores describe behaviour under a named benchmark version; they do not measure intrinsic moral character or establish a universal safety ranking.
Release status: current, code and historical result artefacts public. Next, an Eastwood ARC-Align v0.x candidate replication release. Not claimed: validated standard, contamination-proof benchmark, intrinsic-alignment measure or independently replicated result.
In an author-run March 2026 pilot, an explicit instruction to identify affected people increased AI-judge stakeholder-care scores in five analysable model runs. One model run failed; scoring was nonblind and single-scorer. The result demonstrates prompt responsiveness under that protocol, not embedded values, persistence or validation of a “Stewardship Gene”.
The Fisher-combined p-value is not headlined: independence among component tests is not established.
In Eastwood’s specified model, correction proportional to x^beta dominates modelled drift proportional to x^k asymptotically when beta > k. Separately, Eastwood reports a 45-trajectory author-run mechanism pilot in which conditions differed under a constructed drift protocol. The pilot did not estimate beta or k; its GPT-3.5 engine was evaluated by another OpenAI-family model rather than a cross-family judge; its coded H1 rule differs from the stated positive-slope hypothesis; and it is not an empirical validation of the inequality.
Confirmatory-run status: not ready to submit or run. The inspected v3 harness was unsafe and protocol-mismatched. A synthetic-only rc1 repairs prompt orthogonality, sandbox isolation, fail-closed panels, carried-state identity, HMAC block randomisation, hidden-bank input disjointness, exponent and bootstrap-validity gates, replicate-cluster bootstraps, schedule-aware missingness and the held-out gate; 35 of 35 non-network tests and its synthetic fixture pass. It deliberately has no real provider adapter or collection command. Exact endpoints, evaluator controls, calibrated fresh banks, encrypted commitment, budget, outer image digest, analyst custody and the final freeze remain unresolved. A prospective registration must disclose and exclude prior pilots and govern a new dataset.
Eastwood reports a nine-run, single-model exploratory pilot in which the Eden condition had the highest reported mean entangled-fitness score and fewer reward-hacking events than Babylon under the author-defined protocol. The study is preliminary, single-lab and unreplicated. It is excluded from every external-evidence count.
The Darwin Godel Machine is prior work: Zhang et al. (2025).
The programme publishes adverse and limiting results beside positive ones. The early alpha approximately 2.24 estimate is retracted. The later estimate near 0.49 has wide uncertainty. In Paper IV.d, the grouped Gemini exponent is non-significant, DeepSeek is null under the recomputed v5 Spearman test and GPT remains null. Paper V lost one complete model run and uses nonblind, single-scorer evaluation. Paper X does not estimate beta or k. DGM v4 is a nine-run, one-model pilot without independent replication. These results are part of the evidence record, not footnotes to be hidden.
What would change the status. A frozen and independently replicated result can upgrade the relevant claim. Failure under a frozen rerun, a better-supported alternative explanation or materially earlier prior work can downgrade it. The status changes; the previous record remains visible.
I can draft a skeleton argument for the Court of Appeal. I struggle to open routine post. The framework below is my attempt to describe that architecture honestly, as a hypothesis about minds like mine rather than a certificate about my own.
The Polymathic Neurodivergent Profile is Eastwood’s proposed, falsifiable and presently unvalidated descriptive framework. It is not a diagnosis, recognised syndrome or psychometric category. Eastwood reports adult diagnoses of autism and ADHD. Those diagnoses do not clinically certify all six proposed PNP components. Cross-domain learning, recursive pattern transfer, autonomous synthesis, monotropic attention and institutional misrecognition require separate operationalisation and testing.
Its practical proposition is narrower and well supported: visible capability does not, by itself, establish absence of disability or remove the need to assess reasonable adjustments by their functional effects. The PNP offers a testable account of the cognitive profile associated with this work. Whether it explains the work is the hypothesis under examination. The loop is open by design.
A preserved chat record may establish that a user entered or received specified material on a date if the provider record is authenticated. It does not, without further evidence, establish that the material trained a shared model, entered model weights, was memorised, leaked to another user or caused a later output. A later Gemini output using “Eden Protocol” in a thematically related context may be recorded as a dated model-output echo. The causal route is unknown; no claim of training, memorisation, disclosure or derivation is made.
The point is not to ask you to trust me. The point is to make trust unnecessary. Ten steps, a terminal, and an afternoon.
shasum -a 256 <file> and compare with f0d1f38ffd8546152d9d9d28dc5ec083c16a35858f2c12b63e69db7ed50901ad.09f5b5e156ed96f8883eaf668495fd350898ce62be6294b8f788e0e2d6dcb664.In-browser check: verify byte identity. A match means only that the selected bytes match the published digest.
This may be the most useful section on the page. Every sentence below is a claim I could have made, and chose not to, because the evidence doesn’t reach it. If you catch this site making any of them, that is a defect: report it and it will be corrected in public.
Exact bytes public. Historical dates graded. Prior work credited. Results open to replication. Claims open to correction.