a16z podcast: AI models now exploit vulnerabilities, not just find them
Dylan Ayrey (Truffle Security) and Feross Aboukhadijeh (Socket) join host Joel De La Garza on the a16z podcast to discuss how frontier models have shifted from discovering vulnerabilities to actively exploiting them.
Featured · Dylan Ayrey, Feross Aboukhadijeh, Joel De La Garza
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Cable Management extension for ComfyUI published
- Reddit user shares Star Trek-style clip made with Minimax H3
- Claude Opus 5 praised by student, then fails simple chart task
- Airbnb tests AI-powered search with user-controlled toggle
- Bernie Sanders worries AI is a 'Don't Look Up' situation