ChainThink report, July 17: According to CNBC, Microsoft CEO Satya Nadella criticized Anthropic's usage restrictions on the Fable model during an internal meeting with Copilot engineers, calling its approach to rejecting certain requests "excessively editorialized" and stating that such practices lack rationality.

Anthropic's support page indicates that when users inquire about specific large model development topics, the platform may respond using older version models instead.

Previously, Anthropic stated that new safety measures implemented after Fable 5's relaunch might misclassify more benign requests as content requiring interception.

Microsoft previously announced a $5 billion investment in Anthropic, with Anthropic committing to spend $30 billion on Azure infrastructure.

Some of Microsoft's Copilot products also leverage Anthropic's models, making Nadella's remarks relevant to a key partner and customer base.

Nadella further emphasized that enterprises should be able to develop custom models at lower costs and maintain control over their internal data, rather than being monopolized by a small number of companies over AI Token resources. He stated that it is untenable for only two global companies to own Token capital while others are forced to rent access.