Microsoft Restricts Internal Use of Claude Fable 5 Over Data Privacy Concerns

Technology giant Microsoft is reportedly taking a cautious approach to the use of Anthropic’s latest artificial intelligence model, Claude Fable 5. Although the model is available to external customers through services like GitHub Copilot and Foundry, Microsoft is restricting internal access for its own employees. The move stems from deep concerns regarding Anthropic’s data retention policies, which are deemed risky for the confidentiality of sensitive corporate data. According to a report by The Verge, Microsoft’s legal team is currently conducting an intensive review before granting full access to internal staff. Claude Fable 5 is a next-generation AI model categorised as a Mythos Class, claiming capabilities far exceeding previous Claude models, particularly in high-level tasks such as programming and cybersecurity. However, this power brings concerns; Anthropic acknowledges the model could theoretically be misused to create malware. Despite Anthropic’s claims of implementing strict guardrails, Microsoft remains wary of how user interaction data is managed. This situation underscores a growing conflict between the need for AI companies to monitor model outputs and the absolute data privacy requirements of large corporations. Unlike Fable 5, previous Claude models operated without strict data retention rules, making them easier for Microsoft to approve for internal use. Microsoft’s lawyers are still evaluating whether the efficiency gains offered by Claude Fable 5 outweigh the risk of data leakage, demonstrating that data control remains a decisive factor for major organisations adopting new technology, regardless of its sophistication.