Anthropic is the AI lab behind the Claude models, focused on AI safety.
A chronological archive of observed data — an aggregation of collected primary sources, not predictions.