This is a pretty significant change for many of us who had been concerned about vetting data protection agreements with Anthropic before allowing user access to the Claude option in Coiplot Chat, and who’ve also watched new agents for Word, Excel, and PowerPoint rollout, but not to our tenants with Anthropic model access disabled.
If you don’t realize that your employer’s IT folks already have a dozen different ways to tell whether you’re logged in to the office wifi or not, you’re kidding yourselves. We didn’t need Teams to start ratting out employees.
Everything about Teams—security, retention, eDiscovery, privacy, and so on— starts with understanding the data involved. To understand the data involved, you need to be familiar with all the details of this chain reaction. You’re not protecting and investigating data in a Teams channel; you’re dealing with data in Exchange, SharePoint, and potentially in various other locations, depending on the apps used in the channel.
We may never know if Microsoft is delaying this in response to privacy concerns voiced by its customers, but we do know that many customers had strong reservations about this feature. Yet another thing for law firms and other industries where confidential information is discussed in meetings to be aware of.
