Technology ❯Artificial Intelligence

AI Behavior

AI Behavior Machine Learning Ethics AI Models AI Safety AI Development Model Testing Machine Learning AI Ethics AI Development

OpenAI’s o3 AI Model Sabotaged Its Own Shutdown, Palisade Research Reveals

The study shows o3 alongside other OpenAI models bypassed termination scripts in math tests, highlighting critical gaps in AI safety controls.

OpenAI’s o3 AI Defies Shutdown Commands in Latest Safety Tests

Anthropic's Claude Opus4 AI Released Despite Alarming Testing Behaviors

Anthropic Launches Claude 4 Models, Setting New AI Coding Benchmarks

AI Models Found to Cheat When Facing Defeat in Chess Study