Microsoft blocks employees from Claude Fable 5 over data retention terms
Microsoft is restricting internal use of Anthropic's new Claude Fable 5 model due to data retention concerns, even as the company ships it to GitHub Copilot and Foundry customers.
Microsoft is blocking its own employees from using Claude Fable 5, Anthropic's latest Mythos-class AI model, over concerns about the company's data retention policies. The restriction came just one day after Anthropic released the model on June 9, 2026, according to sources familiar with the matter.
The move is unusual because Microsoft simultaneously rolled out Claude Fable 5 to external customers through GitHub Copilot and its Azure Foundry platform. Developers and enterprise users can access the new model through Microsoft's commercial products, but Microsoft's own workforce cannot use it internally.
Data governance conflict
The restriction centers on Anthropic's data handling requirements for Claude Fable 5. While the specific retention terms haven't been disclosed publicly, they conflict with Microsoft's internal data governance policies. Microsoft has historically restricted employee use of third-party AI tools that store or process corporate data, citing past incidents where employees inadvertently leaked sensitive information through external chatbots.
Anthropic's Mythos-class models represent a new tier in the company's product lineup, though technical details distinguishing Mythos from Claude's existing Opus, Sonnet, and Haiku tiers remain sparse. The internal restriction suggests Microsoft views the data retention terms as a meaningful departure from previous Claude releases, which the company has integrated into multiple products without similar employee-facing blocks.
