Table of Contents
New Capabilities Combat Social Engineering and Impersonation Before, During, and After Teams Meetings

Recently, there’s been some fuss and considerable hype about the threat posed by attackers who represent themselves as IT help desk workers when using federated chat to connect to employees. There are many ways to defend against that threat, including operating a tighter external access allow list to define what tenants are allowed to federate with your tenant.
Impersonation doesn’t just happen in federated chat. Teams meetings can also be infiltrated by unwanted guests, including notetaking bots. Hopefully, you’ve updated meeting policies to block those unwanted visits. Bots are not the only unexpected participants who might turn up in a call, perhaps after using social engineering to persuade an organizer to admit them. Three interesting updates are coming to help combat impersonation in meetings.
Impersonation Protection
The first is Impersonation Protection for Teams meetings, described in message notification MC1530613 (29 October 2026, Microsoft 365 roadmap 573157). This update should be with tenants starting in mid-December 2026. Microsoft says that Teams can detect “potentially deceptive participants and meeting organizers.” I obviously attend very nice meetings and have never seen Teams issue a risk warning about suspicious identities lurking in the participant gallery, but apparently the idea is that Teams will highlight the problem (“I wouldn’t join that meeting if I were you…”) and after that it’s up to users to decide whether to join or leave the meeting.
Obviously, the risk of false positives exists in any situation where computers attempt to detect problems. In this case, the risk is that some perfectly valid external participant, such as an external advisor brought in to consult on some issue, is flagged as problematic. Impersonation protection is a good idea in itself, but it’s likely to be yet another black box where administrators have no visibility into the signals used to make determinations. There’s no word on whether tenants can disable Impersonation protection.
Microsoft doesn’t say what licenses are required to use Impersonation Protection.
Facilitator Scans Your Transcripts
Message center notification MC1536056 (3 November 2026, Microsoft 365 roadmap item 573450) reveals that the Facilitator agent is getting a new protection skill in late January 2027.
In this case, Facilitator uses meeting transcripts to identify things like action items and summaries. The agent can now use its new skill to scan the transcript for evidence that anyone in a meeting is attempting to “trick users into sharing sensitive information.” Users don’t have to instruct Facilitator to look for social engineering tricks. The check happens automatically and cannot be disabled for a specific meeting, meeting organizer, or tenant without disabling Facilitator.
I guess sending a chat message or making a direct request to share something like the worksheet containing next year’s salary data might be an example, but Microsoft doesn’t say what it considers sensitive information to be. One person’s critical document is another’s opportunity to recycle a file. Nor does Microsoft say whether signals such as the presence of a high-priority sensitivity label are considered. Such a label would seem to be a worthwhile signal to use.
If Facilitator detects a problem, “users will see a warning with options to end the meeting or temporarily restrict information sharing.” Cue an embarrassing silence if the warning appears in perfectly reasonable circumstances. Again, this feature is something I suspect that people will take time to get used to, and Microsoft might also need extra time to tune how the agent detects problems. Nevertheless, it’s an interesting example of extending an agent capability with a new skill.
The use of the Facilitator agent requires Microsoft 365 Copilot licenses.
Cleaning Up After a Problem
Last, we have message center notification MC1532241 (30 October 2026), which explains how Microsoft Defender can help track details of problematic calls or meetings. As Microsoft says, “when investigating a suspicious Microsoft Teams call or meeting, your security team needs to understand what happened, who initiated the activity, and the organization associated with the initiator.” You’ll need Microsoft Defender for Office 365 Plan 2 (in E5 SKUs) to use the new capability.
Available now, the CallActivityEvents table in Microsoft Defender XDR advanced hunting holds information about activities in Teams calls and meetings. The idea is that the information about who scheduled calls and what happened could provide valuable information for forensic investigations into behavior like social engineering and attempted impersonation.
An Attractive Target
The size of the Teams installed base inevitably makes it an attractive target for attackers. As stronger multifactor authentication, passkeys, and the retirement of insecure APIs make traditional routes into tenants more difficult, attackers must find other techniques to compromise accounts. Social engineering and impersonation are obvious choices because gaining control of a user account provides an excellent foothold inside a tenant. For that reason, it’s good to see Microsoft adding these capabilities.
Microsoft 365 for IT Pros continues to earn 5-star reviews from readers—and for good reason. If you manage a Microsoft 365 tenant, you’ll benefit from practical, real-world guidance written by experts who work with the platform every day. The subscription includes extensive coverage of Microsoft 365 administration, security, compliance, and governance, plus a 450-page book dedicated to automating Microsoft 365 with PowerShell.