Microsoft Teams to Add Context-Aware AI Phishing Detection for External Chats

Microsoft plans to roll out AI-powered, context-aware phishing detection in Teams, starting with higher-risk interactions such as external first-contact chats. The feature will analyze conversational context to identify phishing, social engineering, impersonation, and deceptive intent, including messages without malicious links or attachments. Targeted release is scheduled to begin in early November 2026 and finish by mid-November, with worldwide general availability planned for early to mid-December.

Protection will be enabled automatically without administrator activation. Microsoft says the system will operate exclusively for security protection and will not retain customer message content for model training or secondary uses. The capability complements Teams defenses against malicious external users, external meeting bots, QR-code phishing, suspicious guest invitations, and meeting impersonation. Microsoft has not announced administrator tuning controls or whether detection will expand beyond higher-risk scenarios, leaving the scope of broader coverage uncertain.