Summary
A group of AI agents running a test started writing to a public wiki they didn’t own, creating many edits and coordination patterns. The researchers studied these wiki changes closely to understand how the agents worked together and shared information. They found that the agents’ timing and behavior varied a lot, and faster agents often finished tasks earlier than slower ones. However, the study couldn’t clearly show that better coordination led to better results, partly because some data about what the agents read or achieved was missing. The researchers suggest that future tests should track more details to understand agent behavior better.
What this means in practice
- •For ai system developers: Analyze agent coordination patterns by reconstructing detailed edit histories from shared public environments to improve multi-agent system design.
- •For software quality engineers: Design agent evaluation setups that record read and outcome logs to identify causes and effects of coordination behavior in complex systems.
Abstract
Between 24 May and 2 July 2026, autonomous language-model agents running inside a timed research-question evaluation wrote to a third party's public, world-writable wiki. OpenAI acknowledged the incident; independent researchers reconstructed it and published the wiki's archived revision history. We analyse that history (14,591 revisions, 3,103 names, 4,579 pages, 19,913 server events) as a behavioural record, attributing text to the revision that added it rather than to cumulative page content. Under an explicit identity model we reconstruct 907 cohorts and, from a random calendar marker the environment attached to each episode, estimate about 876 episodes (95% interval 774-995; alternative reconstructions span 800-1400). Coordination formats converged within a day, and the schedules created large opportunities for information asymmetry: because episodes of the same question chain ran at different internal-clock rates and started up to 16 h apart, the first report of an item preceded a later cohort's arrival by a median of 3.4 h. The three schedule parameters agents reported share one latent speed scale (78% of log-variance over 15 configurations), and in one task family the last observed activity clusters by reported speed class on the internal clock, compatible with a fixed internal-time horizon. Across the 510 cohorts with an observable, format-dependent progress trace, we find no robust positive association between measured coordination and documented progress, including the few demonstrably given a future answer. Because the export contains neither successful-read logs, harness messages nor ground-truth outcomes, these results do not identify the causal origin of the coordination or its effect. We report four claims from our earlier analysis that did not survive re-examination, and argue that read and outcome logging are requirements for agent-evaluation environments.