The AI Observatory now aggregates consented conversations from seven datasets to analyze how people actually use LLMs. This public platform bypasses curated reports from OpenAI and Anthropic. It provides the first independent corroboration of user behavior. Researchers and policymakers can now audit model utility without relying on corporate PR.