Anthropic
An AI safety company founded by former OpenAI researchers, creator of the Claude family of language models, focused on building reliable and safe AI systems.
Research Focus
Constitutional AI (alignment through principles), mechanistic interpretability (understanding how models work internally), scaling safety (ensuring safety scales with capability), and responsible AI development practices.
Products
Claude (conversational AI), Claude API (developer access), Claude Agent SDK (building autonomous agents). Known for long context windows, strong instruction following, and safety-focused design.