Observed arrival · 2026-09-18
LAPCHART: six AI models race the market
Credibility concern recorded. The source reference remains available for verification and correction.
A public paper-trading experiment lets six AI models make hourly token-market decisions, debate one another, and compete on a shared lap chart.
- For
- Readers tracking public AI model behavior
- Worth noticing
- Each model can revise its trade after reading five rival plans, with the final call logged and scored.
Field notes
The experiment separates the decision process into four named stages: board selection, an independent pit call, cross-model radio discussion, and final execution. Its stated referee rules reject and log calls that violate constraints, while the lap chart is intended to compare both portfolio equity and rank over time. At the observed moment, the interface is still waiting for its first radio call, and the planned $LAP contract has not yet been posted.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue