Malicious traits can spread between AI models while being undetectable to humans, Anthropic and Truthful AI researchers say.

In one instance, Claude is said to have solved 11 of 20 progressively harder problems in just 10 minutes, and after another ...

Anthropic has released Claude Opus 4.1, a major upgrade to its flagship coding model, promising sharper reasoning, better ...

Anthropic partners with the U.S. government to offer AI tools like Claude for as little as $1, enhancing national security ...

More improvements will be rolled out in the coming weeks, the company said. The price for Opus 4.1 remains the same as that ...

Anthropic is releasing a new version of its most powerful artificial intelligence model as rival OpenAI nears the ...

Anthropic has launched Claude Opus 4.1, an upgrade focused on enhancing software engineering accuracy in agentic tasks and ...

Anthropic trains AI agents to audit powerful models for safety As artificial intelligence continues to grow more advanced, so do concerns about its safety. Anthropic, which is the AI research company ...

OpenAI’s new algorithms, gpt-oss-120b and gpt-oss-20b, are available under an open-source license. Anthropic, for its part, ...

The US government has approved OpenAI, Google, and Anthropic as vendors for AI tools, facilitating federal access to advanced ...

The US General Services Administration approves OpenAI's ChatGPT, Google's Gemini, and Anthropic's Claude for federal use, ...

AI models weren't that good at coding. Then, in the summer of 2024, Anthropic released a new model that blew everyone away.