Research Scientist in Scalable Oversight

Conduct foundational AI safety research on debate-based oversight using computational complexity, graph theory, or game theory.
- Develop theory on scalable oversight and adversarial sensemaking within AI safety via Debate frameworks.
- Investigate judge-oracle complexity limits, graph properties for honest strategies, and equilibria stability in debate games.
- Hire team members, supervise mentees, and shape research direction for the Scalable Oversight group.
- Requires established research track record in computational complexity, algorithmic information, graph theory, or game theory.
- Requires rigorous thinking, research vision, and ability to set coherent research agendas in a small, fast-moving organization.

Conduct foundational AI safety research on debate-based oversight using computational complexity, graph theory, or game theory.
- Develop theory on scalable oversight and adversarial sensemaking within AI safety via Debate frameworks.
- Investigate judge-oracle complexity limits, graph properties for honest strategies, and equilibria stability in debate games.
- Hire team members, supervise mentees, and shape research direction for the Scalable Oversight group.
- Requires established research track record in computational complexity, algorithmic information, graph theory, or game theory.
- Requires rigorous thinking, research vision, and ability to set coherent research agendas in a small, fast-moving organization.

Iliad is an umbrella organization that advances applied mathematics research in AI alignment through conferences, research incubation, and fellowship programs.