Posts

Why EAs are skeptical about AI Safety 2022-07-18T19:01:37.457Z
Mastermind Groups: A new Peer Support Format to help EAs aim higher 2022-05-31T18:37:14.213Z
Cost-Effectiveness of Air Purifiers against Pollution 2020-07-30T08:37:48.908Z

Comments

Comment by Lukas Trötzmüller (Lukas T) on 🌏 A hack to solve all of humanity's major problems · 2022-08-11T09:58:54.346Z · EA · GW

There is no extreme poverty or starvation in democratic countries

This seems like a strong claim to me. What's your source for that?

and access to education and health care is one hundred percent, at least in older democracies. Younger ones are getting there fast.

Where do you draw the line between older and younger democracies? Isn't the US pretty old compared to other democracies [1] - and does it provide "100% access to health care" to its citizens?

all countries and all people lived in democracies the major problems of humanity would be solved or be dramatically smaller.

Would you classify X-Risks like AI and pandemics as major problems? Do you think having more countries be democratic would reduce these risks - given that the existing democracies don't do enough on either?

[1] https://www.weforum.org/agenda/2019/08/countries-are-the-worlds-oldest-democracies

Comment by Lukas Trötzmüller (Lukas T) on Why does no one care about AI? · 2022-08-07T23:25:45.290Z · EA · GW

pretty much generally agreed upon in the EA community that the development of unaligned AGI is the most pressing problem

While there is significant support for "AI as cause area #1", I know plenty of EAs that do not agree with this. Therefore, "generally agree upon" feels like a bit too strong of a wording to me. See also my post on why EAs are skeptical about AI safety

For viewpoints from professional AI researchers, see Vael Gates interviews with AI researchers on AGI risk.

I mention those pieces not to argue that AI risk is overblown, but rather to shed more light on your question.

Comment by Lukas Trötzmüller (Lukas T) on Leaning into EA Disillusionment · 2022-07-22T05:23:12.893Z · EA · GW

I found myself confused about the quotes, and would have liked to hear a bit more where they came from. Are these verbatim quotes from disillusioned EAs you talked to? Or are they rough reproductions? Or completely made up?

Comment by Lukas Trötzmüller (Lukas T) on Why EAs are skeptical about AI Safety · 2022-07-20T07:28:04.489Z · EA · GW

The sample is biased in many ways: Because of the places where I recruited, interviews that didn't work out because of timezone difference, people who responded too late, etc. I also started recruiting on Reddit and then dropped that in favour of Facebook.

So this should not be used as a representative sample, rather it's an attempt to get a wide variety of arguments.

I did interview some people who are worried about alignment but don't think current approaches are tractable. And quite a few people who are worried about alignment but don't think it should get more resources.

Referring to my two basic questions listed at the top of the post, I had a lot of people say "yes" to (1). So they are worried about alignment. I originally planned to provide statistics on agreement / disagreement on questions 1/2 but it turned out that it's not possible to make a clear distinction between the two questions - most people, when discussing (2) in detail, kept referring back to (1) in complex ways.

Comment by Lukas Trötzmüller (Lukas T) on Why EAs are skeptical about AI Safety · 2022-07-19T09:57:55.098Z · EA · GW

I'm not quite sure I read the first two paragraphs correctly. Are you saying that Cotra, Carlsmith and Bostrom are the best resources but they are not widely recommended? And people mostly read short posts, like those by Eliezer, and those are accessible but might not have the right angle for skeptics?

Comment by Lukas Trötzmüller (Lukas T) on I’m Offering Free Coaching for Software Developers in the EA community · 2022-07-18T19:38:35.092Z · EA · GW

I'm a software entrepreneur transitioning into higher-impact ventures. Had a mentoring call with Yonatan a couple of months ago. What I really liked about his approach was the structure of the call: First, gathering an overview of the issues of the table. Second, going through them in a fast-forward kind of way. And third, figuring out which ones are the most important to talk about.

The outcomes of the call directly led into the next steps I needed to explore this path.

The fact that he asked for feedback at the end of the call shows me that Yonatan is serious about helping EAs and getting better at it. I did have one or two pieces of critical feedback to his style, but these were just minor points that are to be expected for someone without a lot of coaching experience. I can recommend having a call with him without hesitation.

Comment by Lukas Trötzmüller (Lukas T) on Are there English-speaking meetups in Southeast Germany? · 2022-06-13T13:37:12.100Z · EA · GW

The German-speaking EA meetups I know are all very happy to switch to English whenever non-German-speakers are present. Can't imagine it would be a problem anywhere in the German-speaking world!

Comment by Lukas Trötzmüller (Lukas T) on Mastermind Groups: A new Peer Support Format to help EAs aim higher · 2022-06-02T09:49:06.245Z · EA · GW

Would you say that inexperienced people benefit less from a Mastermind than experienced people? Or would you say that they benefit so little that a Mastermind is not worth for them?

If your claim is that Masterminds are only worthwhile for experienced people, then I disagree for two reasons:

First, the way I see Masterminds, one core aspect is that a group of peers can be much more effective in thinking through problems than a single individual. This is true even if none of my peers have any experience that I don't have. It is surely not true for any imaginable group, but I would guess it is true in the case of intellectually humble and reasonably smart people with similar values, who come together explicitely to support each other. I personally have managed to solve dozens of tricky personal & professional problems with the help of Masterminds groups - problems which I was not able to solve on my own.

Second, people with little experience might need more personal support and motivation than people who already have a lot of experience. They also have less access to mentoring. Obviously, this is not always true, but it seems plausible.

Comment by Lukas Trötzmüller (Lukas T) on Jelle Donders's Shortform · 2022-05-02T08:40:47.083Z · EA · GW

I believe a documentary could be a great vehicle to explain EA and get people interested.

Obviously it would need to explain EA principles. But there is also room to include emotion and personal stories. Which might be much more important, in terms of the effect on the viewer.

Perhaps the emotion and personal stories could make up more than half of the film. One documentary that does this really well is "Chasing Ice". It's about James Balog, a photographer documenting climate change by filming glaciers. The film presents the science in a clear way, but it's mostly just a story about one person trying to have an impact with their career. And that has a much bigger effect on the viewer than a pure fact piece. (This film led me on a path to eventually joining EA)

If the goal is to make a documentary that is viewed by a large number of people, then I see a number of challenges:

  1. It would need to be a really good film with a captivating story - not an advertisement, and not an explainer piece.

  2. The quality (both technical as well as artistically) would need to be exceedingly high, and you might need a very experienced team. Or outside help. For example, "Chasing Ice" was made by an inexperienced filmmaker but got help from an Oscar-winning editor.

  3. It might be difficult to get the film distributed to a mass audience. This depends on (1) and (2), but also on the team.

Comment by Lukas Trötzmüller (Lukas T) on There are currently more than 100 open EA-aligned tech jobs · 2022-04-27T11:28:24.670Z · EA · GW

Which roles specifically seem net-negative to you?

Comment by Lukas Trötzmüller (Lukas T) on Research help CO2 offsetting/compensation programs: How to evaluate them for impact? Alternatives to carbon offsetting? etc. · 2022-04-27T11:26:28.074Z · EA · GW

What does your definition of "offsetting" include? Only projects that reduce CO2 in a very direct way (e.g. building clean power plants)?

Or would you include political advocacy and research? If so, check out the work of Founders Pledge:

  1. https://founderspledge.com/funds/climate-change-fund
  2. https://founderspledge.com/research/fp-climate-change
Comment by Lukas Trötzmüller (Lukas T) on Should I fly instead of taking trains? · 2021-12-26T12:38:29.218Z · EA · GW

According to Founders Pledge estimates, the CO2 savings from donating 100 USD (maybe 1 ton per USD, with high uncertainty) will greatly exceed the emissions from your flight (which might be on the order of magnitude of 1 ton) [1]. Donating USD 100 to Atmosfair, while less effective, would also offset this flight [2]. If you include the value of your time, the cost of the train trip might be far, far higher.

Plane emissions are further complicated, if you live in the EU, by emission certificates - which might cause a counterfactual CO2 saving, when deciding not to fly, of almost zero.

I personally stopped flying in 2015 - but not out of EA considerations. I made that decision because it feels good and unlocks new forms of adventure. It also simplifies my life (by removing certain travel options) and opens up conversations with interesting people who pursue the same lifestyle. I also consider it an exercise in personal growth. It is not a good way to help the climate when compared with donations, and I am open to the possibility of flying again in the future.

[1] https://founderspledge.com/research/fp-climate-change

[2] https://www.atmosfair.de/en/offset/fix/

Comment by Lukas Trötzmüller (Lukas T) on Kaleem's Shortform · 2021-12-01T06:55:05.042Z · EA · GW

A new report from Founders Pledge just came out - although it's just an overview article and doesn't go into much depth. https://founderspledge.com/stories/changing-landscape

Comment by Lukas Trötzmüller (Lukas T) on Sleep: effective ways to improve it · 2021-11-16T17:57:41.600Z · EA · GW

Relevant work demonstrating bright full-room lighting as an alternative to an SAD lamp: https://forum.effectivealtruism.org/posts/bwhDhZQvbEcG4FEb8/preprint-is-out-100-000-lumens-to-treat-seasonal-affective

Comment by Lukas Trötzmüller (Lukas T) on Cost-Effectiveness of Air Purifiers against Pollution · 2021-11-02T16:44:17.618Z · EA · GW

Yes, that's what it is. "We" as in "the author and the reader". There is no co-author or organization involved in this.

Comment by Lukas Trötzmüller (Lukas T) on Defusing the mitigation obstruction argument against geoengineering and carbon dioxide removal · 2021-04-29T22:25:26.727Z · EA · GW

I like your post because it puts some more backstory behind an argument that many people usually accept for face value.

I don't quite understand this argument:

If we can geoengineer or capture enough to offset 60% of our emissions in 2030, and then in 2031 we reduce our emissions by 1% (as measured at the smokestack), then the environmental damage will not fall from 40% to 39.6%, it will fall to 39%. So it's still a one-percentage-point change whether or not we do geoengineering and carbon dioxide removal.

This assumes that geoengineering will cause effects equivalent to removing a certain amount of CO2 from the athmosphere. So each additional ton of CO2 still has the same effect. Is that a good model for the effects of geoengineering? I would naively assume that geoengineering could, by blocking sunlight through aerosols, reduce the impact that every ton of CO2 in the athmosphere has. If this is the case, then 39.6% would be accurate and the argument wouldn't work.

Comment by Lukas Trötzmüller (Lukas T) on Defusing the mitigation obstruction argument against geoengineering and carbon dioxide removal · 2021-04-29T22:20:09.452Z · EA · GW

If we can geoengineer or capture enough to offset 60% of our emissions in 2030, and then in 2031 we reduce our emissions by 1% (as measured at the smokestack), then the environmental damage will not fall from 40% to 39.6%, it will fall to 39%. So it's still a one-percentage-point change whether or not we do geoengineering and carbon dioxide removal.

Comment by Lukas Trötzmüller (Lukas T) on Sustainability Fund · 2020-07-20T08:11:17.832Z · EA · GW

relevant: https://forum.effectivealtruism.org/posts/ucjaZFNrebZdZAnb9/impact-investing-is-only-a-good-idea-in-specific and https://forum.effectivealtruism.org/posts/Rkr2W8ADSGwWXfRBF/effective-impact-investing

Comment by Lukas Trötzmüller (Lukas T) on AMA: Elie Hassenfeld, co-founder and CEO of GiveWell · 2020-03-26T20:01:43.672Z · EA · GW

The AMA recording is here: https://www.youtube.com/watch?v=qNfhG6pTZvM

Crowdsourced notes: https://docs.google.com/document/d/1DNPhrj6rfD8C3K2rfNvhP48Z0XaI91aq4McuV_umL2k/edit#heading=h.fnqw4893e1db

Comment by Lukas Trötzmüller (Lukas T) on All Bay Area EA events will be postponed until further notice · 2020-03-09T21:10:41.733Z · EA · GW

The most interesting part of your post, to me, is your risk model [1]. I would be curious to hear some more feedback from other people on it.

I turned it into a Guesstimate, making some adjustments to some of the numbers and using population figures from Austria [2].

[1] https://docs.google.com/document/d/1A0jcxj4n0BvNt_jMunHT5WSsAKFzuVJJyaaqcK9Z1HU/edit#

[2] https://www.getguesstimate.com/models/15367

Comment by Lukas Trötzmüller (Lukas T) on COVID-19 brief for friends and family · 2020-03-01T11:07:20.495Z · EA · GW

Thanks for the explanation on extreme individual precautions, that made things clearer.

I'm curious what you're thinking of when you say "adopt measures that can plausibly be sustained for one year or even longer"?

I'm thinking of simple, low-cost changes to habits and my living environment that reduce chances of infection with Coronavirus and other illnesses. For example: improving personal hygiene practices (how to handle laundry, when to desinfect hands, how to keep the kitchen super clean, desinfecting electronic devices), changing workout times (to hit the gym at times when it's empty), invite my friends to use hand sanitizer when eating together, going to smaller events instead of mass gatherings, keeping a 1-month food supply at home.

These would be easy to sustain for a year or even longer. On the other hand, "not leaving the house" or "cancel all events" incurs significant cost, so I would not be willing to do that for an entire year unless the risk was very high. (Of course, the risk might become "very high" if >10% of the population becomes infected).

Comment by Lukas Trötzmüller (Lukas T) on COVID-19 brief for friends and family · 2020-02-29T09:13:45.063Z · EA · GW

You mention exponential spread, working from home, and avoiding travelling after March.

But what is the endgame here? How long do we need to stop travelling for? Should we apply these measures, as far as possible, starting in April and keep them up until a vaccine is available in 1-2 years? Will the number of cases level off eventually?

I assume there is no scientific consensus on these questions. If the virus is here to stay, then there might be little value in adopting extreme individual precautions for just one or two months. Afterwards, when you stop taking these extreme measures, there will be an even higher risk of infection. Under these circumstances, it seems better to adopt measures that can plausibly be sustained for one year or even longer.

Comment by Lukas Trötzmüller (Lukas T) on Is fundraising through my hobbies an effective use of my time? · 2020-01-25T08:09:55.885Z · EA · GW

I would think hard about what the relevant resources are that you're trading off against each other. Are your hobbies important for your well-being and relaxation? Is it possible that by starting to monetize your hobbies, you might get less enjoyment out of them? Maybe it will also create some imbalance as you spend more time on them than you otherwise would? Or perhaps it's the opposite and monetizing your hobbies would actually increase the quality of your leisure time? Perhaps you can run a time-limited experiment to find out.

Also, as a full-time online entrepreneur my opinion is that neither blogging nor streaming are particularly good income streams unless you have reason to think you would be exceptionally good in either of these.

Also, have you read this? https://www.lesswrong.com/posts/3p3CYauiX8oLjmwRF/purchase-fuzzies-and-utilons-separately

Comment by Lukas Trötzmüller (Lukas T) on EA Hotel Fundraiser 5: Out of runway! · 2019-10-25T18:41:47.585Z · EA · GW

You could use the donation swap system at EAHub: https://donationswap.eahub.org