rosehadshar

Karma: 2,375

rosehadshar 12 Sep 2022 12:15 UTC
27 points
0 ∶ 0
in reply to: Lizka’s comment on: What happens on the average day?
there seems to be a large variance in how comfortable people are with numbers, but I think this is surmountable
Wanting to flag that my background is entirely qualitative, and I spent many years thinking this meant that I couldn’t do things with numbers. I now think this is false, they aren’t magic, and you don’t need to have deep aptitude for maths/technical training/a background in stats to be able to fiddle around with basic numbers in a way that helps you think about things.
What links here?
- Invisible impact loss (and why we can be too error-averse) by Lizka (6 Oct 2022 15:08 UTC; 129 points)

rosehadshar 21 Aug 2020 14:10 UTC
24 points
0 ∶ 0
in reply to: FMD90’s comment on: What FHI’s Research Scholars Programme is like: views from scholars
Impressiveness: good question, but feels hard to express without going into lots of detail, so I’m going to pass.
Acceptance rate: 9/~150, then 10 out of ~250. We’re planning to take 8 in this round. The summer fellowship was 27/~300.
Some support options, briefly:
- Talking with Owen, the programme director
- Talking with me or other future project managers on the programme
- Peer support
- FHI provides opportunities for coaching and other external support
- We have various structures that aim to help people with this, like 6-week project cycles, a major project in the second year, a quarterly review process and an advisory board...
- For the project cycles and major projects scholars would by default have something like a project supervisor
Apologies for the brief response, writing in haste!

rosehadshar 25 Jul 2023 8:39 UTC
15 points
3 ∶ 0
in reply to: kokotajlod’s comment on: Who’s right about inputs to the biological anchors model?
Don’t apologise, think it’s a helpful point!
I agree that the training computation requirements distribution is more subjective and matters more to the eventual output.
I also want to note that while on your view of the compute reqs distribution, the hardware/spending/algorithmic progress inputs are a rounding error, this isn’t true for other views of the compute reqs distribution. E.g. for anyone who does agree with Ajeya on the compute reqs distribution, the XPT hardware/spending/algorithmic progress inputs shift median timelines from ~2050 to ~2090, which is quite consequential. (See here)
For someone like me, who hasn’t thought about the compute reqs distribution properly, I basically agree that this is just an exercise (and in isolation doesn’t show me much about what my timelines should be). But for those who have thought about it, the XPT inputs could either not matter at all (e.g. for you), or matter a lot (e.g. for someone with Ajeya’s compute reqs distribution).

rosehadshar 3 Apr 2022 11:09 UTC
13 points
0 ∶ 0
in reply to: tyleralterman’s comment on: Pre-announcing a contest for critiques and red teaming
I think a list like this might be useful for other purposes too:
- raising the profile of critiques that matter amongst EAs, thus hopefully improving people’s thinking
- signalling that criticism genuinely is welcome and seen as useful

rosehadshar 19 Jul 2023 11:09 UTC
12 points
1 ∶ 0
in reply to: Erich_Grunewald’s comment on: What do XPT forecasts tell us about AI risk?
I wasn’t around when the XPT questions were being set, but I’d guess that you’re right that extinction/catastrophe were chosen because they are easier to operationalise.
On your question about what forecasts on existential risk would have been: I think this is a great question.
FRI actually ran a follow-up project after the XPT to dig into the AI results. One of the things we did in this follow-up project was elicit forecasts on a broader range of outcomes, including some approximations of existential risk. I don’t think I can share the results yet, but we’re aiming to publish them in August!

rosehadshar 6 Dec 2021 9:40 UTC
12 points
0 ∶ 0
on: Can we influence the values of our descendants?
Thanks for the informal post! I really liked it, and probably wouldn’t have read the whole paper.
I have a few thoughts that I’d be curious for your take on.
I’m a bit unsure how generally the papers you’re looking at apply to the broad question of changing values. A few intuitions:
- These papers look at measurable and relatively narrow features of the past, and how far they explain features of the present which are again measurable and relatively narrow. My intuition would be that most of what matters in terms of values is pretty hard to measure, and might not show up in very concrete slices of reality. But maybe this is confused: if there were big fuzzy effects, they would show up in narrow concretisations?
- Naively I’d say that past values and culture have had a huge impact on current values and culture. Many religions have persisted for a long time, there are differences in values between cultures that seem to have been relatively persistent, everyone inherits a bunch of assumptions from their society many of which have been around for a long time… (Below, Lumpyproletariat makes a similar point I think.) If all of my ancestors were Chinese, I expect I would see the world quite differently to how I actually do, with European ancestors. I don’t really understand how intuitions like this relate to your findings. Do you have any ideas?
To progress, we’d need to study the cultural legacy of intentional movements. This is a much harder topic, where quantitative proxies will be hard to come by.
Do you have any ideas about how to make progress on this? E.g. examples of intentional movements that seem worth looking into, proxies that might work...?

rosehadshar 5 Apr 2022 11:59 UTC
11 points
0 ∶ 0
on: Pre-announcing a contest for critiques and red teaming
Minor point on how you communicate the novelty point: I’m slightly worried about people misreading and thinking ‘oh, I have to be super original’, and then either neglecting important unoriginal things like reassessing existing work, or twisting themselves into knots to prove how original they are.
I agree with you that all else equal a new insight is more valuable than one others have already had, but as originality is often over-egged in academia, it might be worth paying attention to how you phrase the novelty criterion in particular.

rosehadshar 6 Dec 2021 11:52 UTC
9 points
0 ∶ 0
on: A case for nonviolent protest
Thanks for this post James! I found it thought provoking.
Overall, I’m still not sure what I make of your claims. There are a few things contributing to this, including:
- The post is long and I read it quickly.
- Probably just inferential distance/different background models and ways of thinking. E.g. I’ve never been involved in a mass movement, I’m a historian by background and would go about these questions in a different way methodologically, etc.
- More specifically, I’m used to thinking of EA as a social movement, and other social movements as potentially useful historical comparisons. This post was an interesting refactoring for me, looking at social movements as a general kind of intervention.
I think the main reason I’m not sure what I think of your claims is that I felt that the post wasn’t crisp/precise enough about which interventions you think it might be good to have more of. Some questions that point towards my unclarity here:
- Is it that you want EAs to try harder to change public opinion? Protests/social movements aren’t the only way to do this—think media campaigns, books etc.
- Or maybe you’re arguing that EAs should cause there to be more protests on issues they care about, because protests are actually quite an effective strategy? Setting up new mass movements isn’t the only way to have more protests—what about liaising with existing movements, or setting up a temporary organising committee for a one off thing.
- Or maybe you want there to be more social movements on issues you care about as you think they are key agents of change? These don’t necessarily need to be mass movements to make a difference, and even mass movements don’t necessarily need to operate centrally via protest....
I could give more examples of my confusions here. Definitely part of what’s happening here is that I am personally confused about how these concepts and phenomenon relate. Another contributor is that I haven’t taken the time to carefully go through your post and construct the clearest most charitable take here. I’m quite sure that if I did this, it would already clear up quite a bit of my confusion. But I also think it’s likely that you could benefit from expressing yourself more crisply here. I think clarity here matters for two main reasons:
1. Making your assumptions transparent. (E.g. I might be sympathetic to the argument that more social movements would be good, but not think that mass social movements are the main type to favour. Being clear about what you think here would help me to spot research gaps to fill.)
2. Making it easier to check whether you’re using the right kind of evidence and reference classes. (E.g. if you specifically think there should be more protests, then evidence about the effectiveness of phone campaigns is probably not very applicable.)
There are also a few more specific points I’d like to raise:
- You write that longtermism might be a suitable candidate for this sort of mass movement, but that you think global poverty may not be relatable enough. Intuitively this seems a bit wild to me: most people I know find it much easier to empathise with e.g. malnourished children who are suffering right now, than with big abstract numbers on how many people may exist or how much they might suffer.
- I’m not sure, but I think I might have different intuitions to you on how likely it is that a given embryonic social movement succeeds. At one point, you say “I believe you could launch a social movement with reasonable chances of success for less than £100K”. This sounds over-optimistic to me, but I suppose it comes down to what you mean by “reasonable chances”. My guess is that the vast majority of potential social movements never get off the ground, so 0.1% doesn’t seem implausible to me as a base rate for success, and >1% does seem a bit implausible to me. In another place, you say “I can confidently say it’s extremely rare to find such driven and talented people who manage to launch a project of this scale without self-imploding due to conflict and governance issues.” So maybe you also think that successful social movements are pretty rare. It also partly depends on what counts as movement success. Feminism has a very large membership, and is widely accepted as a set of values in lots of societies. But on the other hand, women still own something like 10% of global wealth. (Statistic that I vaguely remember. Haven’t checked.) So if you take full gender equality as the metric of ultimate success, feminism is still very far away from succeeding, in spite of its intermediate successes at growing a movement. You can push this to extremes and argue that just about every social movement has failed (e.g. chattel slavery has been legally abolished but modern slavery goes on, so the anti-slavery movement failed). That’s clearly an unhelpful way to define it, but being too inclusive about success (any movement that’s big must be successful) is also clearly a problem.
- Do you know why OP commissioned the report on funding social movements? Or what conclusions they reached on this, if any? Seems relevant.
I’m excited to see what further research comes out of your project. Some particular things I’d be interested to see:
- A literature review of (relevant bits of) social movement studies, if you think any parts of that literature might be generally insightful for EA/LTism. (FWIW, I feel a lot more optimistic about what we can learn from empirical work than theoretical work here.)
- Counterexamples: examples of protest backfiring. Examples of big important changes where protest was largely irrelevant to the outcome.
- (Maybe) A book review of the MAP book, if you think the model is good. (FWIW, I have a strong ish prior that rigid 8 step models are unlikely to be predictive of messy reality. But you know more about it than me, and think it matches the realities you’ve encountered well. If you back the model, it might be useful to write a good book review.)
- (Not exactly part of the research project) A write up of your key learnings and takeaways from working in XR and Animal Rebellion. I found the asides and comments you made in this post interesting, as I don’t speak much to people in those communities, and it would be interesting to understand better how they do what they do.

rosehadshar 31 Aug 2020 13:15 UTC
9 points
0 ∶ 0
in reply to: Caro’s comment on: What FHI’s Research Scholars Programme is like: views from scholars
I think it can be a good fit for either of those groups. Currently most people are more in the academic work category, but we have a few RSPers who are working on more policy engagement style work, and having a fair bit of success.

It’s also worth pointing out that plenty of RSPers don’t fall neatly into either camp:
- Policy people sometimes do academic style things and vice versa
- Lots of RSPers are exploring and haven’t yet narrowed down to ‘I am definitely going to optimise for policy engagement/some other style of work’
- There are other buckets of activities that RSPers do: software development, teaching and mentorship, community building

rosehadshar 12 Aug 2020 12:08 UTC
8 points
0 ∶ 0
in reply to: Davidmanheim’s comment on: What FHI’s Research Scholars Programme is like: views from scholars
Thanks for this! I agree that a lot of the value of RSP won’t become obvious until after the programme (and also want to flag that as our first cohort is only finishing this autumn, it’s still quite uncertain how large this value will be).
At this stage, the best information we have on how things will shape up for scholars after the programme is what our first cohort have lined up to do immediately after the programme—see here.

rosehadshar 12 Sep 2022 15:31 UTC
7 points
0 ∶ 0
in reply to: finm’s comment on: What happens on the average day?
Thanks so much for this! If this is pedantry, I am very pro pedantry :)
I think this makes my ‘Humans launch 5 objects into space’ section sufficiently dubious that I’ve removed it, but pasting here in the context of your comment:
Humans launch 5 objects into space.
It’s only in the last 8 years that the number of objects launched into space each day has exceeded 1.

rosehadshar 18 Aug 2022 8:41 UTC
7 points
0 ∶ 0
in reply to: saulius’s comment on: What happens on the average day?
Yes, you are completely right. I’ve added ‘farmed’ now; thanks for picking this up.

rosehadshar 17 Aug 2022 15:50 UTC
5 points
0 ∶ 0
in reply to: Lorenzo Buonanno’s comment on: What happens on the average day?
Thanks for the link to Saulius’ post; it’s great and I recommend people check it out.
On the trillion wild birds: yeah you’re right, it’s too high—should be 100 billion instead. Thanks for the spot; have changed.
The number is on p. 89 in the supplementary materials—but importantly it’s just aorder of magnitude, rather than a specific estimate. So it’s consistent with Tomasik’s range.

rosehadshar 4 Jan 2021 10:22 UTC
5 points
0 ∶ 0
in reply to: rsturrock’s comment on: What FHI’s Research Scholars Programme is like: views from scholars
Some examples of policy stuff RSPers have done:
- Advising governments directly (including the Czech and UK governments on their COVID-19 response, the Mexican government on its national AI strategy)
- Networking with relevant government individuals, think tanks etc
- Attending relevant conferences, like the Biological Weapons Convention
- Writing papers, written submissions and reports on various policy issues
Some examples of community building related stuff RSPers have done:
- Co-running programmes like the European Summer Program in Rationality (ESPR), FHI’s Summer Research Fellowship, the Human-aligned AI Summer School, and the Epistea Summer Experiment
- Various kinds of advising calls, including lots via EAG/EAGx

rosehadshar 25 Jul 2023 14:29 UTC
4 points
1 ∶ 0
in reply to: spreadlove5683’s comment on: Summary of posts on XPT forecasts on AI risk and timelines
Good question.
There’s a little bit on how to think about the XPT results in relation to other forecasts here (not much). Extrapolating from there to Samotsvety in particular:
- Reasons to favour XPT (superforecaster) forecasts:
  - Larger sample size
  - The forecasts were incentivised (via reciprocal scoring, a bit more detail here)
  - The most accurate XPT forecasters in terms of reciprocal scoring also gave the lowest probabilities on AI risk (and reciprocal scoring accuracy may correlate with actual accuracy)
- Speculative reasons to favour Samotsvety forecasts:
  - (Guessing) They’ve spent longer on average thinking about it
  - (Guessing) They have deeper technical expertise than the XPT superforecasters
I also haven’t looked in detail at the respective resolution criteria, but at first glance the forecasts also seem relatively hard to compare directly. (I agree with you though that the discrepancy is large enough that it suggests a large disagreement were the two groups to forecast the same question—just expect that it will be hard to work out how large.)

rosehadshar 10 Dec 2021 9:19 UTC
4 points
0 ∶ 0
in reply to: MichaelA’s comment on: MichaelA’s Shortform
Some more recent things:
- Mauricio, What Helped the Voiceless? Historical Case Studies (and a shorter version here)
- James Ozden, A case for the effectiveness of protest
Also fwiw, I have read the ACE case studies, and I think that the one on environmentalism is pretty high quality, more so than some of the other things listed here. I’d recommend people interested in working on this stuff to read the environmentalism one.

rosehadshar 17 Jan 2020 15:40 UTC
4 points
0 ∶ 0
in reply to: OllieBase’s comment on: In praise of unhistoric heroism
Strong agree, thanks for pointing this out Ollie

rosehadshar 24 Jul 2023 14:37 UTC
3 points
0 ∶ 0
on: XPT forecasts on (some) biological anchors inputs
My personal take on these forecasts here.

rosehadshar 24 Jul 2023 10:34 UTC
3 points
0 ∶ 0
in reply to: Ulrik Horn’s comment on: What do XPT forecasts tell us about AI risk?
The follow-up project was on AI specifically, so we don’t currently have any data that would allow us to transfer directly to bio and nuclear, alas.

rosehadshar 20 Jan 2023 10:42 UTC
3 points
0 ∶ 0
in reply to: Sanjay’s comment on: What’s going on with ‘crunch time’?
Thanks Sanjay!

I agree that both of your bullet points would be good. I also think that the second one is extremely non-trivial—more like something it would be good to have a research team working on than something I could write a section on in a blog post.
There’s a sense in which there are already research team equivalents working on it, insofar as lots of forecasting efforts relate to p(crunch time soon). But from my vantage point it doesn’t seem like this community has clarity/consensus around what the best indicators of crunch time soon are, or that there are careful analyses of why we should expect those to be good indicators, and that makes me expect that more work is needed.

rosehadshar

Humans launch 5 objects into space.