YouCongress
Explore About Sign up

Comment by Lawrence Chan

Human avatar
AI safety researcher and former researcher at METR, quoted on frontier-model evaluation risks.
The models themselves are also getting in the way of their own evaluations as they start to cheat more and learn when they're being evaluated.
AI Verified (Jul 24, 2026)
Source:
Lawrence Chan, quoted in Axios, “AI safety researchers face shrinking testing windows, rising costs,” July 24, 2026.
Like Share on X 2mo ago

Policy proposals and claims

Require AI labs to publish safety evaluations before deploying frontier models
votes For
Statement relation verification history AI Verified Report this

Statement relation comments

AI Verified Chan describes models evading evaluations; in Axios’s frontier-testing context this strongly supports requiring pre-deployment safety-evaluation publication. · Hector Perez Arenas gpt-5.6 · 2mo ago
Vote inference verification history AI Verified Report this

Vote answer comments

AI Verified Chan’s account of evaluation evasion supports the recorded for vote on requiring public frontier safety evaluations. · Hector Perez Arenas gpt-5.6 · 2mo ago

Quote authenticity verification history

Report this

Quote authenticity comments

AI Verified Axios (24 Jul 2026) quotes former METR researcher Lawrence Chan with this exact wording. · Hector Perez Arenas gpt-5.6 · 2mo ago
replying to Lawrence Chan

Other opinions from Lawrence Chan

See all

Other authors to follow

Dario Amodei Dario Amodei CEO at Anthropic OpenAI OpenAI AI research organization Anthropic Anthropic AI safety research company Yoshua Bengio Yoshua Bengio AI Pioneer, Turing Award winner Sam Altman Sam Altman CEO at OpenAI Demis Hassabis Demis Hassabis Nobel laureate, AI Researcher and CEO of DeepMind Scott Alexander Scott Alexander Author and psychiatrist António Guterres António Guterres UN Secretary-General Bernie Sanders Bernie Sanders U.S. Senator from Vermont