AI signals in email attacks: Boosting detection and threat hunting

AI-generated content is everywhere, which makes AI “tells” noisy and unreliable as a standalone detection method. But that doesn’t mean they’re useless. When used the right way, AI signals can help you hunt faster, validate suspicions, and reinforce detections - especially when attackers are iterating quickly and trading off old indicators for new ones.

In our upcoming session, Luke Wescott and Alex Orleans highlight some of the AI signals that are out there, break down why some large security vendors downplay their usefulness, and discuss where these signals do create real value.

What you’ll learn

• What AI signals actually are, why they’re trending, and how they show up in both benign and malicious email content.
• How to use AI signals for threat hunting and to strengthen detections via signal stacking (with real examples).
• How Sublime layers deterministic rules and AI to reduce analyst work.

Detecting AI signals is helpful, but it’s not a silver bullet. This webinar will show you where these signals break, where they shine, and how to use them responsibly in real-world detection engineering.

Can’t attend live? Register to get the recording sent to your inbox.