# Anthropic **Anthropic** is an AI safety and research company founded by former OpenAI researchers. It developed the Claude family of language models, emphasizing constitutional AI principles, safety, and highly capable, long-context handling. ^overview