Researcher criticizes embedded evaluations as weak pacing substitute

Stephen Casper criticized the METR/Redwood embedded evaluation report on OpenAI, arguing that 'more evaluations' is a weak substitute for stricter pacing measures like hardware controls, and that OpenAI shaped the evaluation scope, leaving key oversights unaddressed.

2026-09-23 ~ 2026-09-23 · 2 related posts

• Casper: 'more evals' is a weak pacing lever, implementations range from safety-washing to real oversight — StephenLCasper · 2026-09-23
• Casper on METR/Redwood's OpenAI report: OpenAI controlled scope, negligence questions left out — StephenLCasper · 2026-09-23