<aside> 💡 What can you expect: Each semester, we offer two curriculums focused spanning topics on AI alignment and governance.

We will host two technical fellowships, they are

• The Intro Track is aimed for students who are interested in learning the big picture, technical preliminaries, and governance problems pertaining to AI safety.
• The Advanced Track is aimed for students who are interested in grinding out the technical details of state-of-the-art methods (DPO, RLHF), foundation models (GPT-4, Gemini, etc) for you to begin reading and research on technical safety areas.
• We will also have research competitions, speaker events, connections to career opportunities, and social events.

This semester, we will also run a reading group led by a group of graduate students (Oliver Liu,  Deqing Fu) and Prof. Willie Neiswanger on topics pertaining to safety (alignment, mechanistic interpretability, etc). The goal of this group if to meet & discuss papers on a regular basis to brainstorm ideas that lead to publications at top conferences. If you're interested in this please fill out the application! For example, here're a list of papers that I will be presenting in our next meeting: Scaling Laws for Associative Memories, and Birth of a Transformer: A Memory Viewpoint

</aside>

Curriculums