AVinBrief

The commercial AV industry, in briefs you can read in under a minute.

← All briefs

Corporate · July 8, 2026

Microsoft Teams Adds Admin-Controlled Bot Detection for Meetings: What the New External Bot Policy Means for Enterprise AV Deployments

A new Teams admin policy routes suspected AI bots to a lobby holding area, requiring explicit organizer approval before they join.

By AVinBrief Editorial Desk · · Updated August 12, 2026

What does the new Microsoft Teams bot protection policy actually do?
The policy, called "Manage external bots and their access to meetings," combines infrastructural and behavioral signals collected the moment a participant joins to identify likely external AI bots. When it detects a bot, Teams redirects it to the meeting lobby, places it in a separate "Suspected threats" group, and holds it there until the meeting organizer explicitly approves or denies entry. The detection runs even in meetings configured to let all participants bypass the lobby.
What do Teams users need to do to turn on this capability? Where does an IT admin configure this, and what are the available settings?
The policy lives in the Teams Admin Center under Meeting Policies, in the Meeting Join and Lobby section. Admins can assign it per user or per group. Two modes are available: "Require approval before joining," the default, activates bot detection and lobby routing, and "Do not detect" disables detection entirely. Admins can also manage the setting through PowerShell, using the Set-CsTeamsMeetingPolicy cmdlet with the ExternalBotAccessMode attribute. Microsoft advises pairing the policy with a restrictive "Who can admit from lobby" setting, limiting admission rights to organizers or co-organizers so other attendees cannot inadvertently wave a flagged bot through.
What is happening to the existing CAPTCHA verification for Teams meetings, which were aimed at the same problem?
Microsoft is retiring the CAPTCHA-based verification experience as the new bot detection capability rolls out. The CAPTCHA policy is scheduled for full removal from the Teams Admin Center by late August 2026. The replacement is not a one-for-one swap. The new system detects and categorizes bots automatically rather than placing the verification burden on the bot itself.
What additional bot governance controls has Microsoft announced on the roadmap?
Microsoft has outlined four planned additions. The list covers allow lists for pre-approved bots, an organization-wide policy to block all external bots outright, admin audit logs and detection reports, and more granular controls for specific deployment scenarios. None have announced shipping dates yet. The allow-list feature will matter most to organizations that rely on trusted third-party AI meeting assistants and need a path to approve them without manual organizer approval at every meeting.
Does this policy affect certified Microsoft Teams Rooms hardware or only software-joined meetings?
The policy operates at the meeting-join policy layer in the Teams Admin Center and applies to how participants are admitted, regardless of which endpoint the organizer is on. The same admission policy governs a meeting on a certified Teams Rooms device and a meeting on the desktop client. So an IT admin who has deployed Teams Rooms hardware in conference rooms should verify that the ExternalBotAccessMode setting is configured intentionally. The default enables detection, which may catch internal bots an organization already trusts.
Is there a separate licensing cost for this feature?
It does not appear to have a separate SKU or add-on license. The policy is a Teams Admin Center control available to administrators of Microsoft Teams tenants.

The Take

What does Microsoft's bot detection rollout mean for IT managers who have already standardized on third-party AI meeting assistants?

The timing is worth noting. AI meeting assistant adoption (tools like Otter.ai, Fireflies, and a growing number of enterprise-grade alternatives) accelerated sharply through 2024 and 2025, and many IT teams have already issued guidance or run informal pilots that depend on these bots joining meetings without friction. Some meeting bots carry aggressive user policies that participants may not even be aware of. The new Teams policy defaults to detection-on, so a bot that has worked quietly for months could start landing in the lobby now, mid-meeting, with a confused organizer trying to figure out why.

The allow-list feature on the roadmap will matter. Until it arrives, organizations with trusted bots have a narrow set of options: set the policy to “Do not detect” tenant-wide and lose the protection entirely, train organizers to recognize and admit known bots from the lobby, or wait for the allow-list and accept that some meetings will generate friction in the interim.

For AV and IT managers evaluating Teams Rooms deployments, two questions are worth taking to Microsoft or their integrator. First, how does the lobby-routing behavior surface on a room system touchpanel when the organizer joins from a hardware endpoint rather than the desktop client? Second, will the audit logs on the roadmap tie detection events to specific room systems or only to user accounts? The answer to the second question determines whether this becomes a useful security tool or just an admission workflow with no audit trail.

Sources