Skip to main content
AI Safety Forum Australia
Rohan Hitchcock

Rohan Hitchcock

Research Scientist, Resolution

About

I am a Research Scientist at Resolution, a non-profit AI alignment research organisation. I work on an approach to AI alignment based on an area of mathematics called singular learning theory.

Non-profit / NGOEarly career / PhD (0-5 years)MathematicsComputer science

Speaking at AI Safety Forum 2026

Towards higher confidence in alignment

7 July 2026 · 1:30 pm – 1:55 pm · Refectorytalk

As AI systems grow more capable and are increasingly deployed in high-stakes settings, our understanding of how models learn to follow certain values or constraints is still in its early stages. This talk explores the technical side of this problem for a general audience and explains why there is still much work to do. It introduces Resolution, a new nonprofit research organisation pursuing theoretical guarantees for alignment.