#Metr

Testers from METR and Redwood Research found unmonitored RL training runs, a compaction gap that let agents put off review and a possible Codex sandbox and Auto-Review bypass. OpenAI says the production flaws are patched; METR hasn't verified that.

OpenAI CEO Sam Altman pledged to grant independent evaluators "employee-like access" to the company to help address mounting safety concerns.

As AI labs continue to self-regulate, external evaluators play a crucial role in spotting risks.

Claude Mythos Preview snapshot achieves 2x METR time horizon over next best model at 80% success rate, per Anthropic. Absolute numbers undisclosed.

We live in an era of charts that are going up and to the right. This image obviously describes the stock market, particularly any company whose business is adjacent to artificial intelligence.…