A new study suggests that the advanced reasoning powering today’s AI models can weaken their safety systems.
OpenAI’s new safety tools are designed to make AI models harder to jailbreak. Instead, they may give users a false sense of security
November 5, 2025
A small amount of bad data can ‘poison’ even the largest AI models, researchers warn
October 14, 2025
Researchers from top AI labs including Google, OpenAI, and Anthropic warn they may be losing the ability to understand advanced AI models
New research shows that scrubbing risky material from AI training data can build safeguards that are harder to bypass — and one…
Recent tests on OpenAI and Anthropic's AI models show their drive for self-preservation.
The company is boosting its safety testing as it anticipates some models will reach its highest risk tier.
OpenAI has open-sourced two AI safety classifiers that let enterprises set their own guardrails. Experts say the move could…
To test the safety and security of AI, hackers have to trick large language models into breaking their own rules. It requires…
OpenAI and Anthropic tested each other's AI models and found that even though reasoning models align better to safety, there are…
