MIRI wants to support research around the world that is aimed at aligning smarter-than-human AI systems with desirable goals. Our research help program encourages students of mathematics, computer science, and formal philosophy to collaborate and consult with our researchers to help steer and inform their work.

In many cases we’ll be able to connect you with a MIRI research fellow to give a tutorial and answer questions, or critique and make suggestions on the work you’re doing. We might also fly you to Berkeley and host you for a couple days, to make communication easier.

Note that because MIRI researcher time is scarce, at this time we can only assist with theses or papers that make direct progress on the lines of AI alignment research currently active at MIRI: logical uncertainty, decision theory, realistic world-models, Vingean reflection, corrigibility, and value learning.