Director of Epoch, an organization investigating the future of Artificial Intelligence.
Currently working on:
I am also one of the coordinators of Riesgos Catastróficos Globales, a Spanish-speaking network of experts working on Global Catastrophic Risks.
I also run Connectome Art, an online art gallery where I host art I made using AI.
My quick take after skimming: I am quite confused about this post.
Of course the VNM theorem IS a coherence theorem.
How... could it not be a coherence theorem?
It tells you that actors following four intuitive properties can be represented as utility maximisers. We can quibble about the properties, but the result sounds important regardless for understanding agency!
The same reasoning could be applied to argue that Arrow's Impossibility Theorem is Not Really About Voting. After all, we are just introducing all these assumptions about what good voting looks like!
Not central to the argument, but I feel someone should be linking here to Garrabrant's rejection of the independence axiom, which is fairly compelling IMO.
Thank you Lizka, this is really good feedback.
I'd personally err towards different subsections rather than different tabs, but glad to see you experimenting to help EA focus on more object level issues!
Here is a write up of the organisation vision one year ago:
Not sure why the link above is not working for you. Here is the link again:
If you want to support work in other contexts, Riesgos Catastróficos Globales is working on improving GCR management in Spain and Latin America.
I believe this project can improve food security in nuclear winter (tropical countries are very promising as last-resort global food producers), biosecurity vigilance (the recent H5N1 episode happened in Spain and there are some easy improvements to biosec in LatAm) and potentially AI policy in Spain.
Funding is very constrained, we currently have runway until May, and each $10k extends the runway by one month.
We are working on a way to receive funds with our new fiscal sponsor, though we can already facilitate a donation if you write to info@riesgoscatastroficosglobales.com.
(disclaimer: I am a co-founder of the org and acting as interim director)
FWIW here are a few pieces of uninformed evidence about Atlas Fellowship. This is scattered, biased and unfair; do not take it seriously.
My impression of the project remains positive, and this is mostly driven by the involvement of Jonas.
On the other hand, from the description on paper I think it's probably less cost effective and more risky than other efforts like Carreras con Impacto or SPARC.
I'd be curious to hear more from the Atlas alumni and staff about how they think the project went/is going however.
This post argues that:
While I'd love for FHI staff to comment and add more context, all of this matches my impressions.
Given this, I stand with the message of the post. Bostrom has been a better researcher than administrator, and it would make sense for him to focus on what he does best. I'd recommend Bostrom and FHI consider having Bostrom step down as director.
Edit: Sean adds a valuable perspective that I highly recommend reading, highlighting Bostrom's contributions to creating a unique research environment. He suggests co-directorship as an alternative to consider to Bostrom stepping down.