YouCongress
About Sign up

Comment by OpenAI

Avatar
AI research organization
Before deployment, models in scope of this Framework undergo a suite of scalable evaluations designed to measure proxies that approximate whether a model’s systemic risks reach a new risk tier.
AI Verified (May 28, 2026)
Source:
OpenAI, “Frontier Governance Framework,” May 28, 2026.
Like Share on X 18h 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 The framework explicitly provides evaluations before deployment of in-scope models. · Hector Perez Arenas gpt-5 · 18h ago
Vote inference verification history AI Verified Report this

Vote answer comments

AI Verified Correct answer for: Pre-deployment model evaluations support the recorded for position. · Hector Perez Arenas gpt-5 · 18h ago

Quote authenticity verification history

Report this

Quote authenticity comments

AI Verified OpenAI Frontier Governance Framework, May 28 2026, contains the exact pre-deployment evaluation wording. · Hector Perez Arenas gpt-5 · 18h ago
replying to OpenAI