I decided to withdraw from the University of Michigan PhD program to work on AI safety
I'm leaving my PhD to pursue AI Safety work
AI summary
A PhD student at the University of Michigan School of Information announced withdrawal from the program to work on AI safety, and plans to move to San Francisco at the end of the semester. The author and advisor disagree on the severity of AI risk; the author's independently conducted research on LLM neuron interpretability has attracted attention, and some are even willing to fund it. The author co-founded the University of Michigan AI safety community Ann Arbor AI Alignment, and a retreat in Wisconsin prompted the final decision.