Microsoft has acknowledged an error causing its AI work assistant, Microsoft 365 Copilot Chat, to access and summarize some users’ confidential emails by mistake.
The tech giant has pushed Microsoft 365 Copilot Chat as a secure way for workplaces and their staff to use its generative AI chatbot. However, a recent issue caused the tool to surface information to some enterprise users from messages stored in their drafts and sent email folders – including those marked as confidential.
Microsoft says it has rolled out an update to fix the issue, and that it “did not provide anyone access to information they weren’t already authorized to see”. However, some experts warned that the speed at which companies compete to add new AI features meant these kinds of mistakes were inevitable.
Copilot Chat can be used within Microsoft programs such as Outlook and Teams, used for emails and chat functions, to get answers to questions or summarize messages. The blunder was first reported by tech news outlet Bleeping Computer, which said it had seen a service alert confirming the issue.
Experts say that this issue highlights the risks of adopting generative AI tools in certain work environments. Nader Henein, data protection and AI governance analyst at Gartner, said “this sort of fumble is unavoidable”, given the frequency of “new and novel AI capabilities” being released.
Cyber-security expert Professor Alan Woodward of the University of Surrey said it showed the importance of making such tools private-by-default and opt-in only. “There will inevitably be bugs in these tools, not least as they advance at break-neck speed, so even though data leakage may not be intentional it will happen,” he told BBC News.
