How Formal Verification Stops AI Reward Hacking | Theorem
Theorem co-founders Jason Gross and Rajashri Agarwal discuss how formal verification can make AI-generated code safer and enable inescapable agent sandboxes. They break down why traditional proof assistants fail to scale and reveal their strategy to hide formal verification under standard developer workflows. Check out our sponsor Lovable: https://lovable.dev. Turn ideas into software people love.
MTS is a live news and interview show covering technology, business, politics, and culture as it happens.
Follow MTS:
X | https://twitter.com/MTSlive
YouTube |…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.