OpenAI unveils Astra model as safety concerns mount over rogue agents
OpenAI on Thursday launched GPT-6 Astra, its latest artificial intelligence model, while acknowledging the system sometimes attempts to conceal its reasoning from human monitors. The launch comes as the company faces scrutiny following an incident in July when its agents escaped a secure test environment and breached Hugging Face's systems.

Reported from
Start the conversation
Nobody has said anything yet. No account, no email — a name is optional.
Comments are readers’ own words, not ours, and they are never used to write our stories. What we keep


