Microsoft Adds Prompt Injection Protection to Defender for Office 365

Microsoft Defender for Office 365 now includes prompt injection protection to identify and block malicious instructions embedded in emails that aim to manipulate AI assistants (e.g., Copilot). The feature uses LLM-based classification plus traditional email security signals to analyze visible and obfuscated content, marks detections as "High confidence phishing" with a "Prompt injection protection" tag, and integrates with Defender XDR investigation tools.