This is a series of projects within the MINT lab, unified by a broad commitment to enabling liberal democratic societies to navigate the transition to powerful AI with their core values intact. This means not just (as everyone now recognises) building in some form of popular sovereignty, but also ensuring that AI systems actively work to protect and advance individuals' fundamental liberal rights. Note: these are all projects that my lab will undertake at some point; the goal is to find researchers who are interested in working on some subset of them, not to cover them all with this fellowship.
About the project
I'd be interested in support on the projects in the attachment
Theory of change
One of the biggest threats to navigating the transition to powerful AI successfully is that AI will erode individual freedom—in part because of the move to increase democratic control over AI systems. This is a series of projects aimed at exploring whether and how frontier AI systems advance or undermine individual freedom. It includes some conceptual work that situates the individual research projects in a broader field.
Your role
Mentees' level of autonomy will be a function of prior experience and demonstrated capability. Core research questions and judgment will be provided by the PI, with the mentee focusing primarily on implementation, but obviously implementation always raises new questions, and the goal is for mentees to develop the taste and judgment needed to answer those independently.
Prerequisites
Most of the projects require solid LLM engineering skills, focused in particular on behavioural evaluations. Mentees will also be expected to make heavy use of coding agents (subscriptions will be provided). Some of the projects would benefit from a background in the social science.s
Location preference
I have recently moved from Australia to the US, and I am not going to be taking meetings outside of work hours for a good few years, in order to counterbalance the many early mornings and late nights while I lived down under.
Application question(s)
What do you think are the appropriate boundaries between when LLMs can, must, and must not be used in AI safety research? Please respond in 200 words. Do not use AI.
About the mentor
Professor at Johns Hopkins School of Government and Policy; a philosopher working on AI Alignment, Governance and Adaptation, and more broadly on the defence, reinvigoration, and redesign of liberal democratic institutions for the AGI transition.