PeterSlattery

• PeterSlatteryApr 1, 2025, 4:15 PM

  2 points

  0 ∶ 0

  in reply to: Angelina Li’s comment on: An­gelina Li’s Shortform

  Do you think that it would be better to just add a helpful or heart emoji to the post instead? I used to leave the same sorts of comments as Ben. These got downvoted occasionally. I interpreted this pattern as being due to people not appreciating these sorts of ‘thank you comments’. When emoji react were added, I therefore switched to emoji reacting, as I felt that this would achieve the same outcomes without creating the ‘noise’ of a ‘thank you comment’. However, I could go back to leaving comments if that seems like a better approach.

• PeterSlatteryMar 28, 2025, 8:41 PM

  4 points

  0 ∶ 0

  on: Third-wave AI safety needs so­ciopoli­ti­cal thinking

  Thanks for this. Is there a place where I can see the sources you are using here?
  
  I am particularly interested in the source for this:
  ”The other graph here is an interesting one. It’s the financial returns to IQ over two cohorts. The blue line is the older cohort, it’s from 50 years ago or something. It’s got some slope. And then the red line is a newer cohort. And that’s a much steeper slope. What that means is basically for every extra IQ point you have, in the new cohort you get about twice as much money on average for that IQ point. ”

• PeterSlatteryMar 24, 2025, 12:07 PM

  11 points

  1 ∶ 0

  on: Aus­trali­ans for AI Safety Launches New Elec­tion Cam­paign — Here’s How You Can Help

  Thanks for your work on this and in the past Luke and Greg! I’ve also signed the letter. I hope it has a positive impact going forward.

Public Com­ment In­vited on Ar­tifi­cial In­tel­li­gence Ac­tion Plan

• PeterSlatteryFeb 23, 2025, 2:02 PM

  2 points

  0 ∶ 0

  in reply to: Ivan Burduk’s comment on: Ivan Bur­duk’s Quick takes

  Yeah, it still seems to have some issues. To fix something, I tried to disconnect then reconnect the sync, and it has not worked since then. Currently talking to support. But very excited about a future when this works!

• PeterSlatteryFeb 17, 2025, 6:07 PM

  2 points

  0 ∶ 0

  in reply to: JoA🔸’s comment on: Why I Think All The Species Of Sig­nifi­cantly De­bated Con­scious­ness Are Con­scious And Suffer Intensely

  Thank you. I missed that in my skim.

• PeterSlatteryJan 6, 2025, 4:20 PM

  5 points

  0 ∶ 0

  on: Idea: Re­pos­i­tory for AI Safety Presentations

  I think it would be useful, and believe that it doesn’t exist. I’d use the resource, and I know several others who also would. Especially, if it was slides and transcripts from good talks. Have you talked to the people behind https://​​aisafety.info/​​? This could be something they could support/​​host.
  

• PeterSlatteryJan 3, 2025, 9:43 PM

  2 points

  0 ∶ 0

  on: Why I Think All The Species Of Sig­nifi­cantly De­bated Con­scious­ness Are Con­scious And Suffer Intensely

  Thank you for this. I only skimmed it, and even that felt unpleasant given the implications. The main thing I wanted to ask is your view on the ethics of eating bivalves? I’m curious because you’ve clearly given a lot of thought to animal suffering.

• PeterSlatteryJan 2, 2025, 7:53 PM

  9 points

  0 ∶ 0

  on: How Much to Give is a Prag­matic Question

  This is extremely relevant for me as I have been thinking a lot about when to start making more serious donations. I discussed some previous blockers here which haven’t been resolved. I am therefore considering commissioning some research (ideally with others).
  
  Broadly I’m interested in better understanding the ‘donators dilemma’: if you give money now, you forego the later opportunity to ‘give better’ due to having improved information, and to ‘give more’ due to passive income. Also to benefit from increased financial security that might enable you to have more direct impact (e.g., by taking a lower paid role that has higher impact, or starting a new initiative).

  I want somebody to systematically review the literature for to capture the different arguments and trade-offs for giving now versus later. Then to create some sort of accessible decision-making tool or process that people like me can use to decide on an appropriate threshold or strategy to have WRT to giving now versus later.
  
  If anyone is also interested in funding this or knows some existing tools, then please let me know.

MIT Fu­tureTech are hiring ‍a Product and Data Vi­su­al­iza­tion De­signer

• PeterSlatteryNov 13, 2024, 2:18 PM

  4 points

  1 ∶ 2

  on: Con­sider donat­ing to who­ever helped you

  I like this idea, but wonder if CEA or another organization should take the lead on running something like this? Making donations to other people or informal groups is interpersonally and logistically complicated. For instance, people will often refuse a donation if offered it (or when their bank account is requested), and taking money from a person may feel like an obligation, or be misinterpreted. It could work better if they instead donate money to an org who allocates it for a person and contacts them to receive it (and donates it if they don’t accept). That organization could also have a database of credible people/​groups doing work and a general donation option for people who just want to fund movement building of a certain type.
  
  More generally, I’d like to see more effort to proactively identify and fund people who are doing good movement/​community building work. Examples: Talent scout type roles at CEA and some sort of community scan (e.g., a question in the annual community survey and post conferences about anyone who was exceptionally helpful) or input process (e.g., a form where you can notify CEA that someone was particularly helpful in a relevant way).
  
  This could also help address certain oversights. For instance, in my experience, someone can have a lot of positive impact in many regional areas without getting noticed or appropriately supported, which can significantly reduce their impact. Catherine Low and Luke Freeman were once examples of such people.

• PeterSlatteryOct 28, 2024, 1:00 PM

  2 points

  0 ∶ 0

  in reply to: Jamie E’s comment on: Bri­tish pub­lic per­cep­tion of ex­is­ten­tial risks

  Ok, I plan to share the PDF, so let me know when it is good to go!

• PeterSlatteryOct 26, 2024, 12:32 PM

  12 points

  1 ∶ 0

  on: Bri­tish pub­lic per­cep­tion of ex­is­ten­tial risks

  Thanks! Do you want this shared more widely?
  
  Note that I estimate that putting these findings on a reasonably nice website (I generally use Webflow) with some nice interactive embedded (Flourish is free to use and very functional at the free tier) would take between 12-48 hours of work. You could probably reuse a lot of the work in the future if you do future waves.
  
  I am also wondering if someone should do a review/​meta-analysis to aggregate public perception of AI and other risks? There are now many surveys and different results, so people would probably value a synthesis.

• PeterSlatteryOct 10, 2024, 12:27 PM

  3 points

  2 ∶ 1

  in reply to: PeterSlattery’s comment on: PeterSlat­tery’s Quick takes

  I am curious to better understand why people disagree here.

• PeterSlatteryOct 10, 2024, 12:04 AM

  26 points

  9 ∶ 2

  on: PeterSlat­tery’s Quick takes

  I really like the vote and discussion thread. In part because I think that aggregating votes will improve our coordination and impact.

  Without trying to unpack my whole model of movement building; I think that the community needs to understand itself (as a collective and as individuals) to have the most impact, and this approach may really help.
  
  EA basically operates on a “wisdom of wise crowds” principle, where individuals base decisions on researchers’ and thinkers’ qualitative data (e.g., forum posts and other outputs)
  
  However, at our current scale, quantitative data is much easier to aggregate and update on.

  For instance, in this case, we are now seeing strong evidence that most people in EA think that AW is undervalued (as seems will be the case) relative to global development. Also who thinks what in relation to that claim and why. This is extremely useful for community and individual decision-making. It would never be captured in the prior system of coordinating via posts and comments.
  
  Many people may/​will act on or reference this information when they seek funding, write response posts, or choose a new career option. In a world without the vote, and just forum posts, these actions might otherwise not occur.
  
  In short, very keen on this and to see more of this.

PeterSlat­tery’s Quick takes

• PeterSlatteryOct 8, 2024, 8:22 PM

  6 points

  1 ∶ 1

  on: Dis­cus­sion thread: An­i­mal Welfare vs. Global Health De­bate Week

  Low confidence, but my intuition is that animal welfare is more neglected and would have a better ROI in terms of suffering reduced.

[Closed] MIT Fu­tureTech are hiring for a Head of Oper­a­tions role

• PeterSlatterySep 18, 2024, 9:08 PM

  31 points

  20 ∶ 8

  on: Is “su­per­hu­man” AI fore­cast­ing BS? Some ex­per­i­ments on the “539″ bot from the Cen­tre for AI Safety

  Thank you for this. I really appreciate this in-depth analysis, but I think it is unnecessarily harsh and critical in points.
  
  E.g., See: Hendrycks has it backwards: In order to have a real, scientific impact, you have to actually prove your thing holds up to the barest of scrutiny. Ideally before making grandiose claims about it, and before pitching it to fucking X. Look, I’m glad that various websites were able to point out the flaws in this paper. But we shouldn’t have had to. Dan Hendrycks and CAIS should have put in a little bit of extra effort, to spare all the rest of us the job of fact checking his shitty research.

• PeterSlatterySep 11, 2024, 6:22 PM

  13 points

  3 ∶ 0

  in reply to: JulianHazell’s comment on: Step­ping down from GWWC: So long, and thanks for all the shrimp

  I will second the claim that Luke is exceptional, even amongst other exceptional people. He has a rare ability to simultaneously be incredibly impressive, warm, humble, caring, hardworking and productive.