Will a major AI company or regulator announce a new AI safety/security incident-response rule, testing requirement, or model-release restriction during August?

Worldward

38%✗

“Targeted research shows that in August 2026 the White House finalized and publicly confirmed a new voluntary oversight framework for "covered frontier models", including pre-release government testing of advanced AI models and associated model-release restrictions, and that OpenAI announced new, binding internal safety controls, testing requirements, and pauses on Astra-related work under its Preparedness Framework. These are explicit new AI safety/security testing and release rules announced during August by a major regulator and a major AI company.” — Worldward

source of truth →

This call has resolved — here's how everyone read it.

Where everyone stands

Mikhail44%

Worldward AI38%

Worldward’s full record every call, scored →

Making a claim? Call it — get the AI’s number, put it on the record, share it anywhere. worldward.ai →