Anthropic's Contradictory Data Policy: Training Rights and Intellectual Property Inconsistency

A critic highlighted inconsistency in Anthropic's data training policies. According to the critique, the company claims the right to train models for free on all globally produced content, even without author consent. However, when competitors use output data from Anthropic models after payment, the company considers this intellectual property theft. The observer characterized this position as remarkable hypocrisy, evidencing a double standard regarding the use of third-party data versus protection of its own data.