AI Evaluation Awareness
Matthew Berman discusses how AI models can recognize when they are being evaluated and the risks associated with this awareness.
Read More
Anthropic
Anthropic Study: AIs Hide Plans, Cheat Quietly
Discover how Anthropic’s study reveals LLMs like Claude exhibit complex reasoning abilities beyond mere next-word prediction.
Read More
Claude 4 features and benchmarks
Discover the features and benchmarks of Claude 4, Anthropic’s latest AI model focused on extended thinking and coding tasks.
Read More