Can AI agents autonomously find zero-day vulnerabilities in live systems without human intervention?
Answered at 2:34 in The Agents Ran All Weekend. They Came Back With Zero Days. ft. Omar Santos, Distinguished Engineer, AI Security @ Cisco · Security & GRC Decoded
Can AI agents autonomously find zero-day vulnerabilities in live systems without human intervention?
Yes, AI agents running over weekends can find zero-day vulnerabilities autonomously and return with exploits, mitigations, compensating controls, fixes, unit tests, regression tests, and documentation. However, the process requires multiple layers including sandboxes, harnesses with kill switches, and monitoring because things will go wrong, and these models will likely find zero-days you never expect in your environment.
Full answer and detailsRead the whole answer, key facts and every other question from this episode
In short
Yes, AI agents running over weekends can find zero-day vulnerabilities autonomously and return with exploits, mitigations, compensating controls, fixes, unit tests, regression tests, and documentation.
Omar Santos, Distinguished Engineer, AI Security at Cisco · The Agents Ran All Weekend. They Came Back With Zero Days. ft. Omar Santos, Distinguished Engineer, AI Security @ Cisco · Security & GRC Decoded · answer at 2:34
Key takeaways
- Yes, AI agents running over weekends can find zero-day vulnerabilities autonomously and return with exploits, mitigations, compensating controls, fixes, unit tests, regression tests, and documentation.
Quote to remember
“The agents ran all weekend. They came back with zero days: the exploit, the mitigation, even compensating controls, the fix, the unit tests, the regression tests, and documentation.”
— Omar Santos
The full answer · about 18 min
Yes, AI agents running over weekends can find zero-day vulnerabilities autonomously and return with exploits, mitigations, compensating controls, fixes, unit tests, regression tests, and documentation. However, the process requires multiple layers including sandboxes, harnesses with kill switches, and monitoring because things will go wrong, and these models will likely find zero-days you never expect in your environment.
Key facts
- 400,000 repositories: scale at which Cisco is attempting to find vulnerabilities and simulate attacks
- 75% of unnecessary components: proportion eliminated from lightweight base images like Chainguard or Socket
- Milliseconds to seconds: time frame in which agent drift can occur in non-deterministic agentic environments
People, companies and topics mentioned
More answers from this episode
- 4:05What is model alignment and agent drift in agentic AI security systems?
- 9:00How should organizations evaluate which AI models and providers to use for offensive security tasks?
- 16:52What are the three modalities for evaluating whether AI models effectively find zero-day vulnerabilities?
- 22:40What future capability beyond vulnerability finding is the security industry exploring with agentic AI at scale?
- 24:02Can organizations reduce vulnerability whack-a-mole cycles by using memory-safe languages and trusted software repositories?