Summary: Researching how to control (“align”) highly-advanced future AI systems is now officially cool, and UC Berkeley is the place to do it.
Interested in AI alignment research? Apply to Berkeley for a PhD or postdoc (deadlines are approaching), or transfer into Berkeley from a PhD or postdoc at another top school. If you get into one of the following programs at Berkeley:
- a PhD program in computer science, mathematics, logic, or statistics, or
- a postdoc specializing in cognitive science, cybersecurity, economics, evolutionary biology, mechanism design, neuroscience, or moral philosophy,
This was not something I could confidently offer you two years ago. Back then, it was pretty hard to find tenured faculty who would accept you as their grad student or postdoc to work on AI alignment. But with recent advances in machine learning, and so much news buzzing about the long-term impact of AI, more and more senior academics are getting interested, and I can help you find them.
You do still have to be a bit of a trailblazer to research AI alignment. There are no “experts” who have 30+ years of experience in it, so if you’re looking to learn a discipline from someone else who’s already mastered it, this field isn’t for you. But if you want to help invent a field, now is the time, and in my humble opinion, Berkeley is the place, because:
- The Center for Human Compatible AI, led by Stuart Russell, is just forming at UC Berkeley;
- The Machine Intelligence Research Institute (where I work) is a 2-minute walk from the Berkeley campus;
- OpenAI and Google Brain are both located in the Bay Area, and
- In the US, Berkeley’s PhD program in general is ranked by the National Research Council as
- #2 in computer science;
- #4 in statistics; and
- #5 in mathematics. (This last point saddens me, because it was #1 just a few years ago… but 5 is still quite good.)
While my time is fairly limited, I care a lot about this field, and you getting into Berkeley is a reasonable filter for taking time away from my own research to help you kickstart yours.
This is not an offer to help you get into Berkeley. There is a lot of interest building in AI alignment right now, and I’m counting on the standard admissions process to help me screen for the most promising researchers to work on it. In particular, I can only mentor so many people, so please try to draw on other resources to help with the get-into-grad-school step. After that, I can help you with the get-into-AI-alignment part.
So, get into the application process, get into Berkeley, solve AI alignment, and make the future super awesome. And if you want any introductions after you complete the “get into Berkeley” step, shoot me an email: firstname.lastname@example.org 🙂
Also, in case you didn’t notice, Berkeley is super beautiful and located in California 😉