Microsoft denies using Office docs for AI training

Microsoft’s privacy practices regarding AI training have come under scrutiny following recent misconceptions about how the company uses data from its Microsoft 365 suite of applications.

Key misunderstanding clarified: Microsoft has explicitly stated that it does not use customer data from Microsoft 365 applications to train its large language models (LLMs).

• The controversy stemmed from confusion over a default privacy setting for “optional connected experiences” in Microsoft Office
• These connected experiences enable features like online picture search and internet-based information lookup
• The setting’s disclosure language did not specifically address AI training, contributing to public confusion

Technical context: The “optional connected experiences” feature represents basic cloud connectivity functions rather than AI training capabilities.

• A Microsoft learning document from October 2024 listing various content analysis features added to the uncertainty
• The Microsoft 365 X account clarified that the setting only enables basic internet-dependent features like document co-authoring
• Frank Shaw, Microsoft’s communications head, also stepped in to address and correct the misconceptions

Industry parallels: Similar privacy concerns have emerged across the tech sector as AI development accelerates.

• Adobe recently faced comparable scrutiny when its terms of service were misinterpreted as allowing AI training on user content
• The company quickly revised its terms to explicitly state this was not the case
• Major tech companies like Meta, X, and Google do opt users into AI training by default, highlighting the contrast with Microsoft’s approach

Broader implications: These incidents reflect growing public awareness and concern about how personal data might be used in AI development.

• Users are increasingly vigilant about their data privacy rights and how their information might be used for AI training
• The misunderstandings highlight the importance of clear communication from tech companies about their data usage practices
• There’s a notable distinction between companies that do and don’t use customer data for AI training, creating potential competitive differentiation in the market