Who is Michael Darius Eastwood?
Michael Darius Eastwood is a researcher and author in London, the founder of Recursive Dynamics and the author of the ARC Theory and the Eden Protocol. His programme tests, in public, whether the safety of AI that improves itself depends on values and correction built into how it reasons (why he is doing this). The ARC Theory states five laws under test about growth, correction and persistence.
Text reviewed 2026-10-31. Any earlier audio edition may differ; the dated text and linked records govern.
The programme's earliest dated record is a self-emailed manuscript of 8 December 2024 (SHA-256 prefix f0d1f38f), whose headers, public on OSF since 15 October 2026, carry a verifiable DKIM signature; it states the embedded-correction alignment thesis in his own words.
He is the author of Infinite Architects: Intelligence, Recursion, and the Creation of Everything, 468 pages, paperback ISBN 978-1806056200 published 2 January 2026 and ebook ISBN 979-8302196163 published 6 January 2026. The published edition runs to 115,439 words and 37 named concepts.
He is principal investigator on the ARC and Eden research programme, OSF DOI 10.17605/OSF.IO/6C5XB, with ORCID 0009-0004-3222-7442 and Wikidata Q140775170. Thirty-six documents sit in its six areas, from theory to governance. Unit 83 (DOI 10.17605/OSF.IO/MN73W), the programme's first registered study to reach a verdict, completed on 28 October 2026: P21 is supported under the registered test, by the programme's own run, on the access axis, family-dependent, with its mechanism axis not evaluable and one call retried under a disclosed recovery amendment (results: DOI 10.17605/OSF.IO/98HKR). He runs his own experiments with six models from six providers, states the defects his own code review found in them, and one retraction stands published against himself.
Paper II reports error-decay estimates of about 0.49 by regression and 0.59 by endpoint calculation, whose bootstrap interval runs from -1.3 to 2.9. The result is inconclusive. These do not directly measure a capability-growth exponent or self-referential coupling. The early single-model estimate of about 2.24 was retracted when a blinded replication on five frontier models did not reproduce it.
Eden Legal AI is a legal working environment for source-linked records, document preparation and checking. Eden Control checks named agent actions and keeps a best-effort record of what happened. Both hosted apps are in private beta. Eden Research is the trading name of Recursive Dynamics Ltd. The papers are free to read and reuse under the Creative Commons Attribution 4.0 licence. The study code and data published with them are open to inspection, all rights reserved.
Two decades building systems in the music industry taught him how recursive amplification behaves in the wild, and his clients reached Rolling Stone, The Times and the BBC.
I built the source-and-review system first for my legal cases and adapted it to research, which follows the same pattern of sources, citations and precedents. Both use a three-valued gate (allow, hold or deny) whose checks run in code before an action is taken or a change is committed.
These checks are mine and internal. They are not independent; they catch only what their rules encode; they cannot decide whether a theory is true; and they do not replace peer review or replication by others, which this page exists to invite.
The papers have yet to be reviewed by a journal, and every result so far comes from his own laboratory: replication elsewhere is still to come (the standard, field by field). How this work is checked