Gaurav Yadav
Lead @ Bristol AI Safety Centre—bristolaisafety.org
Prev. @ CEA/EV.
Yadav
Karma: 673
Hmm I’d very keen to see what an answer to this might look like. I know some people I work with are interested in making a similar kind of switch.
It might be helpful to also think of China’s compute access in a world where they invade Taiwan. I don’t think this should be weighed highly IMO but still seems personally useful to work through.
i’d recommend reading the following for people interested by this—https://www.foreignaffairs.com/china/illusion-chinas-ai-prowess-regulation
I assume the actions you’ve taken can’t be shared? (No pressure if it can’t).
Hmm I disagree. I’m not a fan of making it a norm for someone to reply to me, feels icky and I don’t think anyone has responsibility to message me back unless we’ve scheduled something beforehand.
I empathise with the awkwardness of trying to reach out again but something like ‘Hey I tried reaching you at EAG: London, but didn’t get a response. No pressure but if you’d have time at this EAGx to have a chat I’d love to. Some things I’d like to get out of our conversation: X, Y, Z….’ could be reasonable way of dealing with this.
How long will reaching a Risk Awareness Moment and CHARTS agreement take?
Small note—it seems to occasionally mess up the usernames for posts like ‘EU farmed fish policy reform roadmap’ and ‘New probabilistic simulation tool’.
(Linkpost) Alexander Berger is now the sole CEO of Open Philanthropy
I’m doing some thinking on the prospects for international cooperation on AI safety, particularly potential agreements to slow down risky AI progress like CHARTS. Does anyone know of a good website or resource that summarizes different countries’ current views and policies regarding deliberately slowing AI progress? For example, something laying out which governments seem open to restrictive policies or agreements to constrain the development of advanced AI (like the EU?) versus which ones want to charge full steam ahead, no matter the risks. Or which countries seem undecided or could be persuaded. Basically, I’m looking for something that synthesizes various countries’ attitudes and stated priorities when it comes to potentially regulating the pace of AI advancement, especially policies that could slow the race to AGI. Let me know if you have any suggestions!
I wonder if anyone has examined the pros and cons of protesting against AI labs? I have seen a lot of people uncertain about this. It may be useful to have someone have a post up, having done maybe <10 hours of thinking on this.
Suggestion: Enlarge the font size for pronouns on EA Global/EA retreat name cards
There was a period when I used they/them pronouns and was frequently misgendered at EA events. This likely occurred because I present as male, but regardless, it was a frustrating experience. I often find it difficult to correct people and explicitly mention my preferred pronouns, especially in socially taxing environments like EAGs or retreats. Increasing the size of the pronouns on name cards could be helpful.
(Intro/1) - My Understandings of Mechanistic Interpretability Notebook
Although I think this is incredible, there is a nagging feeling in the back of my mind that it may be difficult to sustain the kind of success rate that CE is currently achieving. Do you anticipate that, as you run more incubation programs, you will see more incubated charities failing or shutting down?
Do you have a sense of whether you will eventually exhaust cost-effective charity ideas as you continue to run more incubation programs?
Thanks for writing this! This is cool to see.
I may have missed this but do you have a sense of whether this marketing push is leading to more people working on pressing problems that 80k endorses? I am curious if there is a direct correlation between money spent on marketing and people working on these problems.
Very Briefly: The CHIPS Act
Yadav’s Quick takes
Shame you couldn’t get involved with stuff here in Bristol; I think we could have made more of an effort last year to be more open to folks outside of the University of Bristol—glad you’re liking it in Oxford!
Yeah, I concur. I feel like this could have been under a separate tag that we could have hidden instead.
I think you intended to link to this instead of the 80K article at the top?
Maybe you’re referring to this—https://forum.effectivealtruism.org/posts/56CHyqoZskFejWgae/ea-is-a-global-community-but-should-it-be?