Leading EA at UPenn. Rising sophomore.
Email: hazemh@seas.upenn.edu.
Hazem Hassanđ¸
Karma: 139
I specified undergrads because I assume theyâre similar to me. This can of course apply to non-undergrads.
Also, I recommend checking prediction markets and prioritizing elections that are close calls (e.g. donât volunteer for a 90%-likely-to-win candidate, or a 10% one that is dwarfed by two other 40% candidates). Aim for 35-50% (judgement call)
More EA undergrads should do political volunteering. Itâs impactful AND fun.
Choose an election thatâs impactful (e.g. AI safety candidate) and neglected (e.g. primaries in always-blue/âred places), couch-crash the weekend there, and volunteer with the campaign.
I say this after doing 15 hours of street canvassing myself. I was surprised by how anecdotally impactful and fun it was. If you like people-watching, talking to strangers, and/âor joining passionate projects for a weekend, I think youâll also love this.
I wish I thought of this earlier.
Literature on the impact (Claude-generated): Kalla & Broockmanâs meta-analysis of 49 field experiments finds zero average persuasive effect in general elections, but effects do show up when voters lack a partisan cue (i.e. primaries and ballot measures). Mann & Haenschen (2024) find mobilization effects (e.g. canvassing) are 33-76% larger in low-attention races than in high-attention ones. Your marginal volunteer hour goes much further in a primary.
Thank you for the correction and apologies for the error.
Iâm actually not sure where 0.75% came from now that I look at it. I wrote this a month ago and cannot remember my state of mind. I probably didnât double-check the number while Claude-source-finding.
Iâm adding a disclaimer at the top of the post and will update the calculator. Thanks again.
Amazing post (loved the image)
To anyone from Darty whoâs reading this and is on the fence: I revived the EA group at UPenn and itâs been one of my best decisions. Itâs easier than you think (& Iâm saying this as a freshman). Shreyaâs very right that there is a lack of EA presence on campuses; if youâre considering organizing and think âsomeone else will prob. do itâ or âIâm not sure Iâm qualifiedâ I fully believe youâre already in the top 10 people who can organize this, and that the most likely outcome is that all 10 donât do anything. Please change that!
It would be more useful to compare AI safety work vs. other longtermist interventions, since itâs unlikely that donations to GiveWell would beat longtermist interventions from a longtermist POV
I agree that would be incredibly useful; maybe Iâll do that next (20% chance). The same model can be used for pandemics and nuclear riskâIâd just need to update (1) P(doom) for each, (2) tractability (for AI, thatâs the âAI safety decreases risk by 7% per doubling of staffâ), and (3) personal contribution. It could be a quick tool for anyone to realize how impactful longtermist careers are and, based on their beliefs about the world and their own ability, choose the career with the highest EV, though Iâd only recommend acting on that comparison if the difference is quite large (my hunch is 5x or higher) given the uncertainty involved.
It would also force people to hold self-consistent beliefs. If thereâs a separate calculator for AI safety and one for biosecurity, someone could claim that non-[x-risk at hand] is much higher than [x-isk at hand] in each case, but that wouldnât be consistent across the two, cause each [x-risk at hand] would factor into the otherâs non-[x-risk at hand]. In other words, it can be used as a tool to calibrate beliefs about existential risks (I think it would do that for me, at least).
The biggest thing missing from the model is the possibility that safety research is net harmful.
This is quite interesting; I hadnât thought of this. Do you think it should be approximated as â% chance that AI safety is actually badâ and âincrease in AI risk per doubling of staffâ? e.g. it would look like this:
90% chance AI safety reduces AI risk, decreasing it by 10% per doubling of staff
10% chance AI safety increases AI risk, increasing it by 10% per doubling of staff
Or is that too rudimentary, you think?
I compared GiveWell to convince people who believe in global dev but are skeptical of AI risk. I could have kept the explanation of why dollars to GiveWell should be discounted and instead said âDonate [smaller yet still big amount] to AI safety /â longtermist solutionsâ which would be equivalent to âDonate $5M to GiveWellâ (assuming my discounting is accurate), but I feared it would sound circular. Some peopleâs natural response would be âI already donât believe AI risk is that big of a deal!â even though the two framings are logically equivalent
Your post reminded me to revive this project from my drafts and publish it:
https://ââforum.effectivealtruism.org/ââposts/âây7BGB3ikhN4RNzAq5/ââwhat-is-the-expected-value-of-working-on-ai-safety-i-ran-the. TL;DR below:I estimate the expected impact of an additional early-career AI safety researcher by combining assumptions about AI risk, tractability, counterfactual replaceability, and population at stake to express the result in GiveWell-equivalent terms. Under what I believe are very conservative inputs, the estimate is on the order of a few million dollars per year in equivalent donations. The results are very sensitive to some unkown parameters, though. Access the calculator/âmodel used here (tweakable to a wide range of beliefs and judgements).
In short, I think the competing earning-to-give number is $5M/âyr (or so), not $80k/âyr.
Oh that was a complete misunderstanding on my part. Thank you
The messages in the Claude conversation are beyond horrifyingâparticularly:
i am HR
i donât want to make a big deal out of this, it doesnât seem like that big of a deal?If 20% of CEA couldnât do the trivial good (report the incident) for months on end, and the most responsible role to handle this (HR) is getting begged by an AI to do the right thing and still not doing it, then a re-evaluation of the entire structure of CEA might be necessary.
This is very unfortunate and shocking.
Unrelated: how did you get access to this conversation? Are shared Claude conversations publicly searchable/âlisted somewhere?
Your post (along with Anthropicâs statement, and Sam Altmanâs tweet) convinced me to switch from ChatGPT Plus to Claude Pro.
I am not particuarly convinced by Bregmanâs reasoning. I think the frequent prompts (by your post, Anthropicâs, and Altmanâs) to consider boycotting ChatGPT is probably what actually convinced me, which is not very rational to best honest.
My ârationalâ justification is that I think moving $20/âmonth from OpenAI to Anthropic is most likely non-negative (and probably positive). Iâm not sure it why I never thought of this before haha
Thank you!
I might be biased (same school, tried something similar in physics), but great work! I recommend promoting the Olympiads nationally once theyâre stable, so more students can benefit. I believe (not super confident about this, though) that the private tutoring market in Egypt can absorb a lot of demand from students wanting to participate in Science Olympiads, so you shouldnât (at least in my opinion) try to train students en masse; just create the incentive. No idea how you can do that, thoughâI assume social media is the cheapest method?
Also, not trying to bring you down, but Egyptâs bureaucracy is hell (you probably know this already). If you manage to launch just one Olympiad and keep it stable for 1-2 years, thatâs fantastic. If you ever feel like youâre hitting too many walls, itâs totally fine to rethink the whole project and invest your time in other (more impactful) ones
Good luck!
Great job starting this! One small correction, though:
Not quite. Check these out:
EA in Arabic
EA at NYU Abu Dhabi
Effective Altruism UAE
EA at American University of Sharjah
Effective Altruism Middle East (currently dead)
Also, there are groups that are not strictly âArabâ but are (semi?)active in the region like Muslims for EA and EA Israel.
Good luck!