OpenAI benches GPT-6.1 Astra for overstepping the mark from www.theregister.com - Articles on 2026-09-29 13:43 (#78Q7A) Turns out teaching an AI to keep going can make it rather bad at knowing when to stop
