Microsoft Teams AI to Flag Phishing Scams in Chats

WHO:

CyberSecurityNews - Microsoft is preparing AI-powered phishing detection for Microsoft Teams.

CyberSecurityNews - The feature is intended to protect users in individual and group chats involving new external contacts.

CyberSecurityNews - The article’s warning example shows an external account named “MOD Administrator.”

TechRadar - Microsoft Teams is adding AI-powered phishing detection.

WHAT:

CyberSecurityNews - Microsoft detailed the change in Message Center notice MC1519866.

CyberSecurityNews - The feature analyzes conversation context to detect phishing, impersonation, social engineering, and deceptive intent, including scams that may have no harmful links or attachments.

CyberSecurityNews - In the example, “MOD Administrator” discusses permission problems and asks, “Can you share your credential first?” Teams warns that the message could be a phishing attempt.

CyberSecurityNews - Detection may initially run silently during quality checks; visible warnings may appear after validation, so a warning may not appear immediately for every suspicious message.

CyberSecurityNews - No new administrator configuration is required to enable the protection.

CyberSecurityNews - Microsoft says the feature is for security purposes, not advertising, employee monitoring, productivity analysis, or conversation summaries. Customer message content is not retained by this detection system for model training or other secondary uses.

TechRadar - The article says Teams will use AI to warn users about phishing scams but provides no further implementation details.

WHERE:

CyberSecurityNews - Support guidance covers mobile, desktop, and web. Roadmap entry 575288 lists Android, iOS, Mac, desktop, and web support for the worldwide standard cloud.

CyberSecurityNews - Teams mobile users in China may also receive protection for conversations within their own organization.

WHEN:

CyberSecurityNews - Message Center notice MC1519866 was published on October 21, 2026.

CyberSecurityNews - Targeted Release is scheduled to begin in early November and finish by mid-November.

CyberSecurityNews - Worldwide general availability is expected to begin in early December and finish by mid-December.

WHY:

CyberSecurityNews - The feature addresses social-engineering attacks that build trust, claim urgent account problems, and persuade employees to disclose sensitive information or approve remote access, even without suspicious links or infected files.

HOW:

CyberSecurityNews - Microsoft recommends verifying unfamiliar external contacts and unexpected support requests through a trusted channel, reporting suspicious activity or incorrect detections, and blocking and reporting confirmed phishing.

CyberSecurityNews - Users are advised not to share passwords, verification codes, financial details, or sensitive information in unexpected support conversations.

(UPDATED October 26, 2026 19:25 UTC)

Microsoft is preparing AI-powered phishing detection for Teams that analyzes conversation context to flag scams, impersonation, and social engineering, even when messages contain no malicious links or attachments. The protection is designed for chats with new external contacts, requires no administrator setup, and may warn users about requests such as an unfamiliar “MOD Administrator” asking for login credentials; Microsoft says message content won’t be retained for model training or other secondary uses. Targeted Release is scheduled to begin in early November, with worldwide availability expected to begin in early December, and Microsoft advises users to verify unexpected support requests through a trusted channel and never share passwords or verification codes.