A team of researchers at Carnegie Mellon University has released PhishGuard, an open source tool that uses machine learning to detect AI-generated phishing emails with 97 percent accuracy. The tool analyzes linguistic patterns, metadata, and behavioral signals that distinguish synthetic text from human-written messages.
PhishGuard integrates with popular email platforms including Microsoft 365 and Google Workspace through standard API connections. The tool processes emails locally, ensuring that sensitive corporate communications never leave the organization's infrastructure.
The project is available on GitHub under an MIT license, and the research team plans to release regular model updates trained on emerging phishing techniques observed in the wild.
