Empirical companion explores looped computation in AI systems, indicating practical deliverables for audits and policies.
This report is the empirical companion to The Ouroboros Thesis: Looped Computation as a System Primitive for AI Systems (v1). It presents a shipped TypeScript implementation of the Ouroboros runtime, three production case studies (A11oy, Sentra, Amaru), and the experimental harness that closes the v1 agenda on adaptive depth, convergence traces, and loop-budget frontiers. The central claim is narrow and operational: when convergence is measurable, the loop trace is not a log but a deliverable that downstream audit, policy compilation, and regulated procurement can consume as evidence.
No takes yet. Share an insight, caveat, or question.
Lutar, Stephen P., Jr., Stephen (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: