Skip to content

Review Resources

Control

Alternatives to Schemers
Reward Seeker Empirics
- Measuring Reward-Seeking via Contrastive Belief Updates (guided)
Trading with AIs

Tracks @ XLab

Courses on AI safety from the Existential Risk Laboratory at the University of Chicago.

Learn

Tracks
Review
Resources

Your account

Account
My classrooms

© 2026 XLab.

AI ControlModule 6: Beyond scheming: seekers and deals

Reward Seeker Empirics

Optional: Supplementary Talk

The talk below is not required. It covers the reward-seeking findings behind the guided paper walkthrough in this section.

Is AI doing the right thing for the wrong reasons? — Apollo Research

Sign in to track progress

PreviousFail safe(r) at alignment by channeling reward-hacking into a "spillway" motivation Next Measuring Reward-Seeking via Contrastive Belief Updates (guided)