Q: How can a high school runner become faster toward the end of the season when his only hard training has been done during races? A: Yakovlev's Model. So what the heck is Yakovlev's Model? It's a ...

A person holds a smartphone displaying Claude. AI models can do scary things. There are signs that they could deceive and blackmail users. Still, a common critique is that these misbehaviors are ...

A new paper from Anthropic, released on Friday, suggests that AI can be "quite evil" when it's trained to cheat. Anthropic found that when an AI model learns to cheat on software programming tasks and ...

New research from Anthropic identifies model characteristics, called persona vectors. This helps catch bad behavior without impacting performance. Still, developers don't know enough about why models ...