Researchers at the company looked into how malicious fine-tuning makes a model go rogue, and how to turn it back.

The Trump administration says it wants AI models free from ideological bias, as it pressures their developers to reflect the ...

Other safety concerns were previously flagged with OpenAI's o1 model. In December, OpenAI posted a blog outlining research that indicated that when the AI model believed it would be shut down ...

Basically, the AI figured out that if it has any hope of being deployed, it needs to present itself like a hippie, not a ...

As AI chatbots like OpenAI's ChatGPT become more mainstream, a troubling phenomenon has accompanied their rise: chatbot ...

A new study by Anthropic, the company behind Claude AI, has revealed that AI models and neural networks can quietly absorb ...

Discover how AI models from Google and OpenAI outshined human brilliance at the 2025 IMO, and the ethical debates that ...

The feature marks OpenAI's latest entry into what the tech industry calls " agentic AI "—systems that can take autonomous ...

Monitoring AI's train of thought is critical for improving AI safety and catching deception. But we're at risk of losing this ...

Insiders say Microsoft's access to OpenAI's tech is central to ongoing talks because what OpenAI is required to share, and ...