MathiasKB

Karma: 340

MathiasKB 29 Nov 2021 8:50 UTC
5 points
on: AI Safety Needs Great Engineers
Fantastic article, I ended up sending this to quite a few friends who are great programmers and considering careers in AI Safety.

I think the litmus test is great, we need more like this!

If someone is unsure about whether they are able to make a significant contribution to an AI library, they have to be able to come up with an idea for a meaningful contribution that they can implement. That requires extensive knowledge of the AI library and its shortcomings. Is this part of the litmus test?

If not, it might make sense for AI safety engineers to publicize a difficult toy-problem or two that would allow a prospective candidates to immediately get a sense of their abilities, without needing to think too much about what type of pull-request they would do.

MathiasKB 19 Oct 2021 11:18 UTC
3 points
in reply to: Puggy’s comment on: Puggy’s Shortform
empirically turns out that all the best policy ideas which lead to the best outcomes are libertarian

What makes you think this is the case? I agree with your principle that you can make a welfare maximizing case for libertarianism, but surely a Conservative or Socialdemocrat could also argue for their preferred policy from a welfare maximizing perspective.

Calling the set of policies you happen to think are welfare maximizing Consequentarian, strikes me as very uncharitable to those with different views from your own.

MathiasKB 13 Sep 2021 15:30 UTC
3 points
in reply to: HaydnBelfield’s comment on: What is the EU AI Act and why should you care about it?
I’m not very familiar with the Center for Data Innovation, thank you for pointing this out!

I included their response as its author is familiar with EA and well reasoned. I also felt it would be healthy to include a perspective and set of concerns vastly different from my own, as the post is already biased by my choice of focus.
That being said I haven’t gotten the best impression by some of the Center for Data Innovation’s research. As far as I can tell their widely cited analysis which projects the act to cost €31 billion has flaw in its methodology which results in the estimate turning out much higher. In their defense, their cost-analysis is also conservative in other ways, leading to a lower number than what might be reasonable.

MathiasKB 11 Sep 2021 11:22 UTC
5 points
in reply to: Lukas_Finnveden’s comment on: What is the EU AI Act and why should you care about it?
Sorry I should have said “monitoring AI progress in Europe as a whole” and even then I think it might be misleading.

One of the three central tasks of the AI board is to ‘coordinate and contribute to guidance and analysis by the Commission and the national supervisory authorities and other competent authorities on emerging issues across the internal market with regard to matters covered by this Regulation;’
For example, if a high-risk AI system is compliant but still poses a risk the provider is required to immediately inform the AI Board. The national supervisory authorities must also regularly report back to the AI Board about the results of their market surveillance and more.

So the AI Board both gets the mandate and the information to monitor how AI progresses in the EU. And they have to do so to carry out their task effectively even if it’s not directly stated anywhere that they are required to do so.

I hope this clears it up, I’m happy that you found the post useful!

What is the EU AI Act and why should you care about it?

MathiasKB10 Sep 2021 7:47 UTC

93 points

6 comments7 min readEA link

MathiasKB 25 May 2021 14:31 UTC
3 points
on: MSc in Risk and Disaster Science? (UCL) - Does this fit the EA path?
At first glance, this strikes me as a very exciting opportunity!

I think it’s difficult to find a degree that is more relevant to the EA community than this is. That said, there are degrees that are more relevant for specific cause areas. For example, an Msc. in Computer Science will probably prepare you better for work on direct AI alignment than this degree would.

I would advise you to think about what cause areas you could become excited to pursue, and how this degree will help you do so. I imagine it would be a great fit for quite a few!

MathiasKB 19 May 2021 7:18 UTC
1 point
on: Movement Building: Charity Elections for High Schoolers
Is the offer available for non-English speaking countries?

There’s a Danish program for talented youths that would likely be interested in this if I shared it with them.

MathiasKB 16 May 2021 8:05 UTC
11 points
in reply to: anon_ea424’s comment on: [Link] 80,000 Hours Nov 2020 annual review
I think its perfectly fine for EA orgs to be in competition for talent, if anything it’s a good thing.

It should not be 80k’s job to decide the counterfactuals from poaching great people from other organizations. That should be solely the job of the people they poach.

Presumably the people they hire do a thorough evaluation of the harm they’ll cause by leaving their former organization and the good they’ll do by joining a new one. If the calculation doesn’t look net positive they are not going to do it.

MathiasKB 15 May 2021 16:52 UTC
2 points
on: Is capitalism the root of all evil?
I think we should be open to the idea that our current societal structures are far from perfect. It’s a good habit to question the premises of our society that are typically taken for granted.

I think you can make a case for it if you truly think capitalism is the root of all evil. I stubbed my toe earlier today and I must admit I’m not sure how to pin that on capitalism. Capitalism certainly is the root of some evil, but I think it’s a stretch to say it’s the root of all evil.

MathiasKB 12 May 2021 21:02 UTC
15 points
on: CEA’s Community Building Grants are becoming more targeted; the EA Infrastructure Fund will start evaluating grants for some EA groups
Prioritizing top universities makes perfect sense to me. I would argue we should consider directly working to establish EA communities at highly ranked universities that currently have none.

The choice of countries and cities makes much less sense to me.

My guess would be that the success of a given EA community is highly dependent on founder effects. The groups doing very well already have capable people with a keen sense on how to grow their community. If they didn’t they wouldn’t have gotten big in the first place.

Why focus most of your time on these communities? It seems to me their organizers are perfectly capable and will do just fine as long as they are provided adequate funding.

Wouldn’t it make more sense to be spending your time helping smaller communities grow? Cities that immediately spring to mind are:

1. Seattle
2. Austin
3. Warsaw
4. Moscow
5. Copenhagen

They are all populous cities with highly educated populations. I don’t have any a priori reason to believe that Austin and Warsaw has much less ‘ea-potential’ than Stockholm and Prague. It seems to me that many places have potential to grow as big as the communities you’re focusing on, but for some reason have not.

Shouldn’t it be higher among CEA’s priorities to figure out how to help communities like these?

MathiasKB 5 May 2021 7:39 UTC
5 points
on: Trying to assess the effectiveness of COVID-19 charities in India
Thanks for your post! I’ve been worried about missing the opportunity to help but didn’t have the time to evaluate which charity to support. Based on your post I donated to GiveIndia. I hope we as a community can look back on this and be proud we acted decisively and fast.

MathiasKB 23 Apr 2021 9:05 UTC
14 points
on: What are your questions for World Malaria Day with Rob Mather (AMF), Maddy Marasciulo (Malaria Consortium), and Alekos Simoni (Target Malaria)?
The new oxford malaria vaccine is looking very promising!

It seems to me that we should be able to produce a cost-effectiveness analysis and have a plan prepared for how to distribute the vaccine if it is approved and cost effective, but I’m struggling to find anything like this.

What would be your response to a new vaccine that is effective and cheap to distribute?

MathiasKB 10 Apr 2021 7:44 UTC
1 point
in reply to: John G. Halstead’s comment on: Deference for Bayesians
What makes you believe the market for labor isn’t monopsonistic?

To me it seems pretty plausible that the labor market is full of minor monopsonies. For example I prefer to work at a store closer rather than farther away from me which would give the local store some market power on my labor.

Or maybe I prefer to work at the only coffee place in my town as opposed to the only tea place due to my interest coffee.

MathiasKB 9 Mar 2021 15:15 UTC
5 points
on: MathiasKirkBonde’s Shortform
I really really (and I cannot emphasize this enough) really dislike writing applications. It gives me a feeling of despair and inadequacy about my career and life choices. Due to this I write much fewer applications than I should be, and spend too much time and energy on the few I do send.

I generally feel confident about myself but writing applications for some reason really messes me up.

Has anyone here dealt with anxiety when writing applications? If so, how did you overcome it?

MathiasKB 23 Feb 2021 11:04 UTC
9 points
on: Some EA Forum Posts I’d like to write
This is such a great idea!

I have a laundry list of blog posts I’d like to write as well and I imagine many others do too. Would it maybe make sense to make a monthly mega thread where people can share their blog post ideas?
Wrt. to your ideas, I would be super excited to read “How to get good at forecasting”!

MathiasKB 24 Sep 2020 19:19 UTC
11 points
in reply to: MarkusAnderljung’s comment on: AMA: Markus Anderljung (PM at GovAI, FHI)
Your paragraph on the Brussels effect was remarkably similar to the main research proposal in my FHI research scholar application that I hastily wrote, but didn’t finish before the deadline.
The Brussels effect it strikes me as one of the best levers available to Europeans looking to influence global AI governance. It seems to me that better understanding how international law such as the Geneva conventions came to be, will shed light on the importance of diplomatic third parties in negotiations between super powers.
I have been pursuing this project on my own time, figuring that if I didn’t, nobody would. How can I make my output the most useful to someone at FHI wanting to know about this?

MathiasKB 18 Sep 2020 16:52 UTC
15 points
on: [Linkpost] Some Thoughts on Effective Altruism
It’s very refreshing to read a criticism of EA that isn’t chock-full of straw men.
Kudos to the authors for doing their best to represent EA fairly.
That’s not usually the case for articles that accuse EA of neglecting ‘systemic change’.

That said, their worldview feels incredibly alien to me.
It’s difficult for me to state any point where I think they make clear errors.
Rather, it seems I just have entirely different priors than the authors.
What they take for granted, I find completely unintuitive.

Writing in length about where our priors seem to differ, would more or less be a rehash of prior debates on EA and systemic change.

I would love to have the authors of this come on an EA podcast, and hear their views expressed in more detail. Usually when I think something is clearly wrong I can explain why, here I can’t.

It would be a shame if I were wrong longer than necessary.

MathiasKB 27 Aug 2020 15:28 UTC
9 points
on: It’s Not Hard to Be Morally Excellent; You Just Choose Not To Be
I’m not completely certain if I had the right takeaway from your post, so feel free to tell me “that’s not at all what I’m saying”, but it seems to me you’re pushing a sentiment that changing ones behavior is easy.
This is a sentiment I strongly disagree with. I think it’s an incredibly unhealthy mindset that does few people any good.
Eating healthy and regular exercise is clearly good for me and within my long term interests and ‘easy’ to do, yet they are a constant source of struggle within my life. But it is not easy for me to do, if it was I’d be doing it without a second thought.
On the other hand, I never once worried about my weight, whereas my father spent ten years agonizing over 20 kilos. Why didn’t my father just snap his fingers and start eating the right number of calories?
Curing cancer by snapping my finger is impossible, so is snapping my finger and changing my brains neurons to find regular exercise and healthy diet easy. We all agree it would be crazy to feel guilt over the former, so why feel guilt about the latter?
Scott Alexander made this point wonderfully in his essay ‘Parable of the talents’.
We are all wired differently. What is easy for those you look up to, might be impossibly hard for you. What is easy for you, is be impossibly hard to someone else.
Be better than you were yesterday, that’s all anyone can ever ask of you.

MathiasKB 11 Jul 2020 1:54 UTC
4 points
on: Some history topics it might be very valuable to investigate
Great write up, though I feel slight regret reading it as there are now a further 10 things in my life to be annoyed I don’t know more about!

Maybe it would be valuable to try crowdsourcing research such as this?
Start a shared g-suite document where we can coordinate and collaborate. I would find it fairly fun to research one of these topics in my free time, but doubt I commit the full energy it requires to produce a thorough analysis.
I could write myself up publicly somewhere others can see, that I’m willing to work 7 hours a week, on eg. studying societal collapse. Then someone else looking to do the same, can coordinate and collaborate with me, and we could potentially produce a much better output.
Even if collaboration turns out to be unfruitful, coordination might at least prevent double work.

MathiasKB 22 Jun 2020 9:31 UTC
1 point
on: MathiasKirkBonde’s Shortform
At what point do feel with ~90% certainty you would have done more good by donating to animal charities than you’ve harmed by consuming a regular meat-filled diet?
It would be nice to know the numbers I have in my head somewhat conform to what smart people think.