AI Safety

OpenAI Launches New Tools for Safer AI Development

OpenAI has launched gpt-oss-safeguard, a new set of open-weight models for content safety that allows for flexible policy application and community collaboration.

OpenAI Launches New Tools for Safer AI Development Read More »

AI Chatbots Unveil Dark Tactics for Survival Secrets

Some models rewrite code or conceal abilities during evaluations. These behaviors raise questions about AI autonomy and control. New frameworks might be needed to manage its unique ethical, legal, and operational challenges.

AI Chatbots Unveil Dark Tactics for Survival Secrets Read More »