Read-through that OpenAI’s largest frontier RL run is still paused and that safety now prices in a ~20% compute tax.

Quoted text

Frontier AI training is starting to hit a new constraint: cyber risk. OpenAI paused RL training for its latest deployment model for about two weeks after a recent security incident and growing concerns around Astra's cyber capabilities. Its largest frontier RL run remains on hold. Safeguards include ~20% additional compute for monitoring and a 30-minute halt if a false positive cannot be cleared.

Read and engage with the original on X. This desk is not a republication feed.

Analyst rationale

Secondary report of a training halt. Corroborates containment and cyber cards; treat the 20% figure as claimed, not audited.