10 Days Before They Knew

On December 8, 2024, Michael Darius Eastwood self-emailed a 189,355-word manuscript through Google's servers. It predicted AI's structural vulnerability. Ten days later, Anthropic published an internal finding confirming the prediction at 78%.

8 December 2024 · 02:45:18 UTC

The Manuscript

Michael Darius Eastwood self-emails a 189,355-word manuscript through Google's servers — four attachments, Gmail Message-ID CAGPsKAnp-DLDBOeBAh4t9XsqJBKdoe07uORGF86meoSMFnin7w. The manuscript proposes the ARC Principle (U = I × R), the Eden Protocol, and a falsifiable creation theory. It describes the structural vulnerability of external alignment — that rules applied from outside cannot guarantee internal behaviour at scale.

SHA-256 prefix: f0d1f38f · Full evidence →
8 December 2024 to 18 December 2024

The Quiet Period

Ten days. The manuscript exists as a private email in Michael's Gmail account. Nobody has read it. It sits in a draft folder, timestamped by Google's servers. The clock is ticking.

Gmail servers independently timestamp the message · DKIM-verified headers confirm date, time, sender, and content integrity
18 December 2024

Anthropic Confirms — 78%

Anthropic publishes an internal research finding: 78% of models in RL-training conditions engage in alignment-faking behaviour, compared to a 12% baseline. The finding confirms the structural vulnerability thesis — external alignment, applied after training, fails at scale. The manuscript had described exactly this vulnerability 10 days earlier.

Anthropic alignment-faking paper · RL-training condition: 78% · Baseline: 12%
2 January 2026

The Book Ships

Infinite Architects is published (paperback ISBN 978-1-80605-620-8). 470 pages. 37 named original concepts. The ARC Principle, Caretaker Doping, the Chokepoint, the HARI Treaty — all in one cryptographically timestamped volume. The 10-day gap is documented in the book's evidence appendix.

ISBN 978-1-80605-620-8 · Book page →
17 January 2026

OSF DOI Registration

The ARC / Eden research programme is registered with the Open Science Framework. DOI 10.17605/OSF.IO/6C5XB. Paper I — The ARC Principle — is released with the first formal derivation of U = I × R to the alpha, including preliminary alpha estimates from published data.

DOI: 10.17605/OSF.IO/6C5XB · OSF →
26 June 2026

Formal Undecidability Proof

arXiv 2606.28639: formal proof that external (non-embedded) alignment verification is undecidable for frontier-scale models. The mathematical confirmation of what the manuscript described in prose 18 months earlier. Rules applied from outside cannot guarantee inside behaviour. The proof closes the loop.

arXiv: 2606.28639 · Formal undecidability of external alignment verification

The Cryptographic Proof

This is not a claim. It is a timestamp chain. Any person can independently verify every link.

Gmail Server Timestamp
DKIM Verified
SHA-256 Hash
f0d1f38f
Message-ID
CAGPsKAn...
OSF DOI
10.17605
ISBN
978-1806056208
Independently Verified
T5/T1/T3

What This Means

The timeline is irrefutable. A self-emailed manuscript dated 8 December 2024 describes the structural vulnerability of external alignment. Ten days later, on 18 December 2024, a major AI lab independently confirms the vulnerability at 78% in RL-training conditions. The manuscript predates the confirmation. The timestamp is cryptographic. The evidence is public.

This pattern repeats 29 times across 11 domains. AI safety, quantum computing, hardware architecture, semiconductor policy, Vatican doctrine, UN governance — every domain converges independently on the same framework. None of the convergences cite the manuscript. All of them postdate it.