I am a Boston-based physicist working on AI safety because I think advanced AI will be a transformative technology soon, and I want to make sure that transformation goes well. My current interests are on inner alignment (“How can we ensure the AI does what we ask?") and interpretability of neural networks. My work on these topics is available on the AI Alignment Forum.
Previously I was a Research Fellow at the Flatiron Institute’s Center for Computational Astrophysics (2019-22) and before that a Postdoctoral Scholar at the Kavli Institute for Theoretical Physics at UCSB (2018-19). I hold a BS in Physics from Caltech (2015) and a PhD in Astronomy from the University of Cambridge (2018).
[publications] [orcid] [cv]
I sometimes write thoughts I have on space exploration, workflows, agency, mental health, and more here. If you’d like to receive updates you can either use RSS or get posts by email by subscribing here.
In my spare time I like to write science fiction short stories. You can find some of my stories on Github and in the Cambridge University Science Fiction Society magazine [1,2]. I also paint, mostly acrylics, and you can find some of my paintings on my DeviantArt page.
If you think we could make something together, have questions about things I’ve done, or have feedback on things I could do better, please contact me at adamjermyn[at]gmail.com. If you prefer to leave anonymous feedback, you may do so here.