a16z Podcast: Defending Against AI That Can Hack

a16z Podcast · rss · 2026-08-18

a16z podcast discusses new cybersecurity challenges posed by AI agents capable of finding and exploiting vulnerabilities. Experts cover why guardrails can hinder defenders, the need for multi-model access, and the new endpoint security risks created by autonomous agents. They also analyze the struggles of static signatures and honeypots against reasoning software.

Original post →

More from Safety

Safety channel →