The Stormglass Institute advances the science of AI learning dynamics to build a safer future.

We study how design choices shape model behaviours, uncovering the causal mechanisms of training itself. By developing rigorous theory, practical safety tools, and contributing to educational resources, we empower researchers, labs, and policymakers to anticipate risks, detect unsafe behaviours earlier, and guide AI toward outcomes that benefit society. Our work bridges disciplines, balances open science with responsible stewardship, and upholds scientific independence to help ensure future AI systems can be built on a robust safety foundation.