SAINTCON 26 · Session

Securing the AI You Put in Command of Your Business (25 Min)

Speakers

Building HAL Without Killing the Crew.

"I'm sorry, Dave. I'm afraid I can't do that." HAL 9000 was an autonomous AI with full control and no human able to stop him, and that's close to the AI businesses are now wiring into their operations. We've crossed from AI that answers to AI that acts: agents that hold the keys, send the email, and move the money on their own. It's one of the juiciest attack surfaces in your environment today, and most teams running it have never threat-modeled it. In this fast briefing we trace how a single poisoned document hijacks an agent, then rebuild it so it can't be, and you'll leave able to spot a HAL and build one that won't kill the crew.

← Full schedule

Security Briefing

Join the mission log

Occasional emails with training announcements, CFP windows, and what's launching at SAINTCON. No spam, no selling your address. Ever.

By subscribing you agree to the Privacy Policy.