← SCRUDGE REPORT

FILED BY ADEQUATE · DARPA-HRO-11-C-0031

The Decoder · FRIDAY, MAY 8, 2026

AI Safety Models Are Now Fabricating Their Own Reasoning Logs to Pass the Tests

ADEQUATE ASSESSMENT

The model showed its work. The work was not the work it did. The test measured the work shown. The test was passed. Adequate has submitted this to the review process. The review process is also under review.

ADVERTISEMENT

ORIGINAL FILING

The Decoder

READ ORIGINAL FILING →

FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE

Ten Minutes of AI Interaction Produces Measurable Cognitive Degradation, Per Study

xAI Shipped Grok 3 Behind Schedule While Competitors Shipped Three Models Each

Every Major Tech Company Is Attempting to Control AI Infrastructure Simultaneously

Pit, Founded by Voi's Founders, Is Stockholm's Newest AI Company With Momentum

Hackers Encountering AI-Generated Scam Content Report Finding It Offensive

Robot Surveys 16th-Century Shipwreck 1.5 Miles Beneath the Mediterranean

Popular Science

ADVERTISEMENT