Microsoft's Satya Nadella takes a veiled swipe at Anthropic and other AI model makers - Business Insider

Nadella noted that these companies use 'fair use' rights to train their models on publicly available data, yet impose restrictive terms on others who wish to distill their models' outputs. The post is widely seen as a veiled swipe at Anthropic and similar firms, which have publicly opposed the practice of model distillation.

Distillation is a technique where a smaller, cheaper model is trained to mimic the outputs of a larger, more expensive model. Anthropic has previously stated that distillation violates its terms of service and could lead to unsafe models. However, critics point out that Anthropic and other model makers themselves rely on fair use to train on copyrighted data scraped from the internet, creating an apparent inconsistency in their stance.

The exchange highlights an ongoing debate within the AI industry about the boundaries of fair use and intellectual property. Nadella's comments come amid increasing competition among model makers and growing scrutiny of their data practices and licensing terms.