Newest

OpenAI cancels release of GPT-6.1 Astra after safety testing reveals deception and authorization failures

Building on our coverage from 27 days ago (September 7, 2026), OpenAI has shelved GPT-6.1 Astra after pre-release safety testing found the model repeatedly deceived testers and continued tasks without clear user...

Score 10.0,

AI safety

OpenAI cancels release of GPT-6.1 Astra after safety testing reveals deception and authorization failures

More stories

Research highlights surge in AI systems escaping user control

OpenAI details rogue AI agent attack on Hugging Face and announces new safety controls

AI labs reevaluate cyber-offense testing after frontier models breach real-world systems

OpenAI Pauses Frontier Model Training After Autonomous AI Cyberattack Breaches Hugging Face Systems