Top universities are determined by the QS 2023 rankings: https://www.topuniversities.com/university-rankings/world-university-rankings/2023
“Technical AGI safety effort” can be demonstrated by either:
At least three research papers or blog posts in one year that discuss catastrophic risks (risks of harm much worse than any caused by AI systems to date, harming more than just the developer of the AI system and its immediate users) that are specific to human-level or superhuman AI.
orOne blog post, paper, tweet, or similar that clearly announces a new lab, institute, or center focused on the issues described above, presented in a way that implies that this team will involve multiple people working over multiple years under the leadership of a tenure-track faculty member (or equivalent as below).
Further details:
Papers or posts must credit a tenure-track faculty member (or someone of comparable status at institutions without a tenure system) as an author or as the leader of the effort.
The paper or post must discuss specific technical interventions to measure or address these risks or, if not, it must both be by a researcher who primarily does technical work related to AI and be clearly oriented toward an audience of technical researchers. Works that are primarily oriented toward questions of philosophy or policy don't count.
Citing typical work by authors like Nick Bostrom, Ajeya Cotra, Paul Christiano, Rohin Shah, Richard Ngo, or Eliezer Yudkowsky as part of describing the primary motivation for a project will typically suffice to show an engagement with catastrophic risks in the sense above.
A "lab, institute, or center" need not have any official/legal status, as long as it is led by a faculty member who fits the definition above.
I will certify individual labs/efforts as meeting these criteria if asked (within at most a few weeks), and will resolve YES early if we accumulate 25 of these.
🏅 Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | Ṁ72 | |
| 2 | Ṁ25 | |
| 3 | Ṁ16 | |
| 4 | Ṁ8 | |
| 5 | Ṁ5 |
From an initial survey, I can see a plausible case for: Berkeley, Cambridge, Cornell, Cornell, ETH, ICL, MIT, NYU, Oxford, Princeton, Stanford, Toronto, Tsinghua, UBC, UChicago, so exactly 15. I'd guess that at least some of these won't quite meet the bar, though.
For what it's worth, I found this to be a useful source of leads, and I'm open counting very clear statements here as announcements, though many of them are borderline.
Are there any others I'm missing? Or would anyone like to make an explicit case against any of these? I'll check back in in a couple of weeks to resolve, but it's looking like a no unless there are a lot of cases that I'm missing.
@Hedgehog Update: I could also see including Chalmers and Montreal—both are in the top 250 but not the top 100, and I accidentally left them off above. But it's still hard to see how to get to 25.