Should EA influence governments to enact more effective interventions? 2022-09-01T17:48:25.796Z
Did OpenPhil ever publish their in-depth review of their three-year OpenAI grant? 2022-07-04T14:11:04.415Z
The Carbon Almanac – an opportunity for EA climate organisations 2022-04-09T12:01:19.170Z
Are some countries far more suitable for setting up a new nonprofit? 2022-03-31T10:08:54.587Z
Common technical needs of EA organisations 2022-02-17T23:33:08.983Z
Altruistic Agency – free tech expertise for effective altruists 2022-01-12T15:21:08.634Z
Would free tech support and development be useful to your EA organisation? 2021-10-09T16:39:04.681Z


Comment by Markus Amalthea Magnuson (peppersghost) on EA forum content might be declining in quality. Here are some possible mechanisms. · 2022-09-25T08:19:45.375Z · EA · GW

I don’t have a view on that, but it would be cool if it was available as a forum setting (”Weight votes by account age”) and some people might like it better that way.

Comment by Markus Amalthea Magnuson (peppersghost) on EA forum content might be declining in quality. Here are some possible mechanisms. · 2022-09-25T00:08:56.769Z · EA · GW

I wrote this in 2013, might be of interest to those concerned:

Comment by Markus Amalthea Magnuson (peppersghost) on What are people's thoughts on working for DeepMind as a general software engineer? · 2022-09-23T21:54:00.146Z · EA · GW

Why would you want to help build AGI?

Comment by Markus Amalthea Magnuson (peppersghost) on So, I Want to Be a "Thinkfluencer" · 2022-08-15T07:58:26.811Z · EA · GW

I plan to post my reports on LessWrong and the Effective Altruism forum

Why would posting mainly in these tiny communities be the best approach? First, I think these communities are already far more familiar with the topics you plan to publish on than the average reader. Second, they are – as I said – tiny. If you want to be a public intellectual, I think you should publish where public intellectuals generally publish. This is usually a combination of books, magazines, journals, and your own platforms (e.g. personal website/blog, social media etc.)

You could probably improve on your plan by making a much more in-depth analysis of what your exact goals are and what your exact audiences are. It seems to me a few steps are missing in this statement:

I believe such people provide considerable value to the world (and specifically to the project of improving the world).

What would probably be useful is, in a sense, a theory of change on how doing the things you want to do lead to the outcomes you want.

If you do decide to go ahead with this plan, I would also focus a lot on this part:

In contrast, I am quite below average on conscientiousness and related traits like diligence, perseverance, willpower, "work ethic", etc.

You are going to need those in the massively competitive landscape you aim for.

Comment by Markus Amalthea Magnuson (peppersghost) on Why does no one care about AI? · 2022-08-09T13:25:13.544Z · EA · GW

If you speak to a stranger about your worries of unaligned AI, they'll think you're insane (and watch too many sci-fi films).

I'm not so sure this is true. In my own experience, a correct explanation of the problem with unaligned AI makes sense to almost everyone with some minimum of reasoning skill. Although this is anecdotal, I would not be surprised if an actual survey among "strangers" would show this too.

Commenting on your general point, I think the reason is that most people's sense of when AGI could plausibly happen is "in the far future", which makes it psychologically unremarkable at this point.

Something like extinction (even if literal extinction is unlikely) from climate change, although possibly further off in time, might feel closer to a lot of people because climate change is already killing people.

Comment by Markus Amalthea Magnuson (peppersghost) on Grantees: how do you structure your finances & career? · 2022-08-05T01:51:36.052Z · EA · GW

You would essentially be a freelancer. Using that framing instead, there are plenty of resources out there on how to build a life as a freelancer. For an EA-specific perspective, here’s a good starting point:

Comment by Markus Amalthea Magnuson (peppersghost) on Effective Altruism for Muslims · 2022-07-28T00:02:00.298Z · EA · GW

This is very exciting and has huge potential. Please get in touch with the Altruistic Agency for tech needs (e.g. website) when you are at that point, I'd love to help.

Comment by Markus Amalthea Magnuson (peppersghost) on EA for dumb people? · 2022-07-11T14:18:00.155Z · EA · GW

Since sociology is probably an underrepresented degree in effective altruism, maybe you can consider it a comparative advantage rather than "the wrong degree". The way I see it, EA could use a lot more sociological inquiry.

Comment by Markus Amalthea Magnuson (peppersghost) on Did OpenPhil ever publish their in-depth review of their three-year OpenAI grant? · 2022-07-04T22:44:20.964Z · EA · GW

I'm aware of this, and it raises more questions than it answers, frankly. For example, I wonder what the terms were when what was originally a grant to a non-profit, turned into (?) an investment in a for-profit.

Comment by Markus Amalthea Magnuson (peppersghost) on EA organizations should be more coordinated · 2022-06-17T12:35:07.208Z · EA · GW

Anti Entropy are doing a lot of work towards this in the operations area, especially for new organisations. I think a lot of the things you ask for (especially in infrastructure) is currently provided ad hoc and informally (e.g. in various invite-only Slack workspaces) or by service providers and (EA) agencies that charge for it.

Comment by Markus Amalthea Magnuson (peppersghost) on Nick Bostrom - Sommar i P1 Radio Show · 2022-06-13T08:08:12.455Z · EA · GW

Well done! If anyone would ever be interested, Nick has the transcript in Swedish here:

Comment by Markus Amalthea Magnuson (peppersghost) on The dangers of high salaries within EA organisations · 2022-06-10T09:16:07.653Z · EA · GW

More views from two days ago:

I especially recommend this comment:

What would be truly useful is annual (anonymous) salary statistics among EA organisations, to be able to actually observe the numbers and reason about them.

Comment by Markus Amalthea Magnuson (peppersghost) on Misercordia Launch!!! · 2022-06-09T22:08:19.560Z · EA · GW

Great designs! I just have to ask… the "typo" is intentional, right?

Comment by Markus Amalthea Magnuson (peppersghost) on Guided Consumption: Funding Charities by Leveraging Consumer Sentiment in the Broader Economy · 2022-06-09T20:45:38.578Z · EA · GW

I mentioned this in an email to you, but thought I'd leave it in a comment here as well just to make other readers aware of the initiative: BOAS does something similar, in the niche of sustainable baby and kids products. They started out fairly recently and have already donated 2,000 EUR to effective charities. I will check if Vincent who founded BOAS has an account here on the forum, and if so, ask him to make some general comments in this thread, on his experience.

Comment by Markus Amalthea Magnuson (peppersghost) on Jobs at EA-organizations are overpaid, here is why · 2022-06-09T09:58:55.171Z · EA · GW

A concrete follow-up question (anyone, feel free to answer it):

What do you think is the correct salary for some common roles, and why that number?

Comment by Markus Amalthea Magnuson (peppersghost) on Jobs at EA-organizations are overpaid, here is why · 2022-06-09T09:52:30.231Z · EA · GW

Non-profit work is systematically underpaid (and often unpaid) in relation to its value (e.g. for humanity) since market mechanisms are laughably unfit to price it properly. I think the EA community is in a great position to counteract this through a culture of high salaries, good benefits etc. and should use that opportunity. I'm happy you bring it up though, as I think there should be far more research on things like salary, incentives etc. instead of just relying on "business/market common sense" which is only fully appropriate in businesses proper, which most EA organisations are not.

Comment by Markus Amalthea Magnuson (peppersghost) on [Linkpost] Towards Ineffective Altruism · 2022-05-24T09:37:41.989Z · EA · GW

Exactly! Somewhat of a sidenote but I find it relevant: I've seen this thing with many political parties in Sweden that usually have a youth organisation that for various reasons often represents a more radical version of the so-called party line on various issues. Political opponents will try to hold the party responsible for what the youth branch says and does, but it's generally understood by most (I think) that the latter is the avant-garde and should not be conflated with the general views of e.g. those voting for the party in elections. Denying there are important connections between the two would be dishonest, but so would saying they are the same be.

Comment by Markus Amalthea Magnuson (peppersghost) on Hiring: How to do it better · 2022-05-23T14:51:43.987Z · EA · GW

It should probably not even be called a work sample under the circumstances I describe, but rather just work.

For example, if I'm hiring a communicator, I could ask them to spend two hours on improving the text of a web page. That could be a typical actual work task at some point, but this "work sample" also creates immediate value. If the improvements are good, they could be published regardless of whether that person is hired or not. This is also why you would pay an applicant for those two hours.

A very simplified example, but I hope the point comes across. And like you mention, for some types of work such isolated tasks are much more prevalent.

Comment by Markus Amalthea Magnuson (peppersghost) on Hiring: How to do it better · 2022-05-23T14:44:55.126Z · EA · GW

Thank you for an excellent reply. I've for a long time found the "mastery, autonomy, purpose" concept useful and think of it as true – for lack of a better word. That these three aspects determine drive/motivation/happiness to a large extent, in a work context.

Comment by Markus Amalthea Magnuson (peppersghost) on [Linkpost] Towards Ineffective Altruism · 2022-05-23T12:42:01.713Z · EA · GW

It seems to me some criticisms, including this one, paint a picture that does not very accurately describe what most effective altruists are up to in a practical sense. You could get the idea that EA is 10,000 people waking up every day thinking about esoteric aspects of AI safety, actively avoiding any other current issues regardless of scale.

In reality, a fair chunk (probably a vast majority?) do what most would perceive as "traditional" charity work, e.g. working at an organisation that tries to alleviate poverty or promote animal welfare, organising their community (university etc.) to promote doing good, doing research on effective methods for solving large problems in society today, or getting more people and organisations to donate money to charitable causes.

I have a hard time believing the general public actually thinks existential risk research on things like pandemic preparation/prevention is a bad idea or not money well spent. But if you equal existential risk with AI threat, it's a whole other framing.

Every movement will have far-out elements that might be hard to make sense of without a lot of context, but that are also just one facet of the movement as a whole. A lot of the recent criticisms of EA I've seen target longtermism in its most "extreme" form, and drag all of effective altruism with it. The criticism of longtermism is very healthy and useful, in my opinion, but this conflation is concerning.

Comment by Markus Amalthea Magnuson (peppersghost) on Hiring: How to do it better · 2022-05-23T12:19:07.799Z · EA · GW

Very good post. Would love to see more summaries of research on hiring but also what makes employees happy, and similar topics. A question about work samples: What are your thoughts, and what is the research, on using real (paid) work instead of work samples? Meaning, identifying some existing task that actually needs to be done, rather than coming up with an "artificial" one.

Comment by Markus Amalthea Magnuson (peppersghost) on What would you desire in an EA dating site? · 2022-04-30T11:58:58.768Z · EA · GW

You could probably make this happen faster and at a much higher quality by instead spending your time on finding funding (e.g. from the EA Infrastructure Fund) to pay an experienced full-stack developer to build the first prototype, under your guidance. Unless it's also your main career plan to work in web development, then you'll learn a lot from doing the project on your own. (This is from someone who has worked in web development for 15+ years.)

Comment by Markus Amalthea Magnuson (peppersghost) on It could be useful if someone ran a copyediting service · 2022-04-30T11:53:36.463Z · EA · GW

I plan to start offering this – among other things – for free through the Altruistic Agency later this year.

Comment by Markus Amalthea Magnuson (peppersghost) on Most effective way to make a UN-based website? · 2022-04-22T09:30:39.597Z · EA · GW

Why do you need a legal entity? Maybe you could describe the project/website in more concrete terms – that would make it easier to give concrete recommendations.

Comment by Markus Amalthea Magnuson (peppersghost) on FTX/CEA - show us your numbers! · 2022-04-18T12:31:30.569Z · EA · GW

Just a list of projects and organisations FTX has funded would be beneficial and probably much less time-consuming to produce. Some of the things you mention could be deducted from that, and it would also help in evaluating current project ideas and how likely they are to get funding from FTX at some point.

Comment by Markus Amalthea Magnuson (peppersghost) on Making Community Building a more attractive career path · 2022-04-09T19:11:55.045Z · EA · GW

Or test titles on an appropriate audience. For example, come up with 20 different titles, then pay 100 professional recruiters 20 dollars each to vote on which one sounds most impressive. Actually, maybe something like that could be done on an even larger scale to find out how this "career capital issue" can be improved for many EA job roles.

Comment by Markus Amalthea Magnuson (peppersghost) on Making Community Building a more attractive career path · 2022-04-07T22:47:34.373Z · EA · GW

Work to give community builders more job security. How to do this would obviously depend on the situation at hand, but in cases where community builders are on grants one could consider longer grant periods, longer exit grants and/or support and incentives for people to start legal entities which then employ them as community builders.

I wonder if this could be, more or less, a single organisation with the main purpose of providing stability for every single community builder at once. A rather small operations team could probably streamline things like contracts, payroll etc. As you mention, for things like mortgages, having what banks consider "normal employment" can be quite important. But the mental relief of not having to think about a bunch of things is a nice perk too, that employees at "normal" organisations take for granted.

Comment by Markus Amalthea Magnuson (peppersghost) on How could Twitter be tweaked to promote more rational conversations, now that Elon is on the board? · 2022-04-06T22:15:50.682Z · EA · GW

Twitter, as a fairly open platform, could be immensely valuable (because social networks are valuable) for humanity if it was more geared towards its users.

My sense is that people have felt for years that Twitter experiment with features that no one asked for rather than making it a nicer place for existing users. (Not sure how true this is, although I'd personally agree.) This is often noticeable in e.g. this way: (See thread for interesting comments on how Twitter product design is an echo chamber of toxic positivity.)

If a social network of Twitter's size and popularity could be slowly shifted towards an open feature development process informed by its users, it could start to grow into something else than what its most frequent users endearingly call a "hellscape". At some point, it would be so usable and useful that its users might even consider paying (a small fee) to keep their account or create a new one. With such a change, the bot problem would probably be significantly reduced automatically. By then, a Stack Overflow-style reputation system is in place that allows self-moderation of the entire platform. Maybe it ends up truly enjoyable again.

Nothing about the above is too hard for capitalists to stomach, making it much more realistic than asking for hyper-niche features that only EAs and rationalists would like/use. Whatever you want to do to Twitter, there has to be a business case for it, sadly.

Yes, I rant about this regularly for some strange reason that still eludes me. On Twitter.

Comment by Markus Amalthea Magnuson (peppersghost) on EA Forum feature suggestion thread · 2022-04-05T10:44:31.478Z · EA · GW

I would like the ability to sort search results by date. Often, I want to know who mentioned something in the past few weeks or months, this is currently not possible. (I'm guessing the current sort order is by magic, but there is nothing indicating this.)

Comment by Markus Amalthea Magnuson (peppersghost) on Operations is really demanding · 2022-03-31T08:22:28.255Z · EA · GW

I can't think of any specific links or such but I can tell you more: I may not go so far as to say it's the norm in Sweden, but it's definitely common to have an annual "utvecklingssamtal" (personal development discussion) and it's often combined with salary negotiations. Personally, I think these two discussions should be separate.

Good organisations use this opportunity to gather a lot more knowledge than what is related to performance. In particular, it can be a way to have an open-ended discussion about the work environment and what improvements can be made. For example, improvements in the physical environment (it's too cold, too dark, chairs and desks are annoying etc.) or improvements in mental environment (Monday meetings are too long, people ping on Slack too often, my colleague is always late for meetings etc.)

Usually, you would fill out some form with prepared questions beforehand and try to standardise this to use over several years to get a sense of improvements made, i.e. asking the exact same questions every year and to everyone. In other words, combine an open-ended part with a very structured part.

If it would help you a lot more than the above, I could probably dig up some old such documents from previous employers, although I would have to paraphrase them rather than share them directly due to privacy and/or intellectual property reasons.

Comment by Markus Amalthea Magnuson (peppersghost) on Operations is really demanding · 2022-03-29T13:51:42.923Z · EA · GW

Having seen overworked operations staff in several organisations throughout my career, reducing stress and having a healthy culture seem to be key improvement factors regardless of organisation size. (This goes for many roles.) If you consistently can't accomplish everything you need to accomplish in 8 hrs/day – given a full-time situation – you are clearly understaffed and this should be resolved ASAP. There are many other stress reducers, such as many weeks of paid vacation per year, great salaries, clear areas of responsibility, structured interviews on the work environment (not the same as performance reviews!) etc.

It seems like many "normal" organisations are under the delusion that they need to operate under the, literally, military conditions you describe. This "get it done yesterday" mentality can kill the morale in any organisation and especially operations people will take the hit, because they are expected to tie all the bits together. What you describe as "never really quite off-the-clock" is super-dangerous and leads to burnout.

If you have worked in different organisations with vastly different cultures on these issues, it seems wild that any organisation wouldn't prioritise the well-being of their employees, when it so obviously also improves the quality of the work.

A common excuse is that some roles or types of work "are just like that" but when people doing that work start talking to others doing it elsewhere it often turns out not to be the case. It's a matter of culture. I know this from experience in software engineering – one company I worked at had a "no death marches" rule to explicitly counteract a common unhealthy bit of culture at many software companies.

Comment by Markus Amalthea Magnuson (peppersghost) on Pre-announcing a contest for critiques and red teaming · 2022-03-25T19:17:49.361Z · EA · GW

How do you plan to encourage participants outside the EA community?

Comment by Markus Amalthea Magnuson (peppersghost) on How To Become A Professional Software Developer · 2022-03-25T17:56:37.164Z · EA · GW

How come you do not mention open source projects? I don’t know how valuable it is nowadays, but working on e.g. Firefox early in my career definitely helped me learn fast from very good programmers in a real project used by millions. It has been a good CV item as well.

Comment by Markus Amalthea Magnuson (peppersghost) on Announcing the Future Fund · 2022-03-22T13:32:48.202Z · EA · GW

It seems like the Expression of Interest form does not send a confirmation email, unlike some (all?) of the other forms.

Comment by Markus Amalthea Magnuson (peppersghost) on Legal support for EA orgs - useful? · 2022-03-17T14:39:01.512Z · EA · GW

I'm 100% behind this idea, maybe not surprisingly so since I founded the Altruistic Agency, that you even mention in the post :) I think the agency model works exceptionally well for work that is only needed every now and then (makes no sense to hire someone full-time) but is so specific/domain-specific – or might even require a license, like in this case – that no current in-house person can do it. The most common option is to hire contractors/freelancers. An agency can be as simple as acting as a layer in between, a contact point with an already built-up roster of great and vetted contractors.

In my view, and I obviously think about this every day, especially regarding the long-term plans for my own organisation, agencies in the EA community should be much more ambitious than that. Having a permanent agency, ideally with full-time employees, allows you to think on a higher level and in a longer term about how you can improve the entire EA community. A lot of work will be one-off projects, day to day types of tasks, but this kind of work helps build up a profile of what EA organisations in general need. Once you've worked with dozens of organisations, you have quality data that allows you to also start thinking about things like original projects, reusable work, publishing learnings, making recommendations etc.

In short, starting an "X Agency" for EA organisations will allow you to confidently answer the question "What X needs do EA organisations have?" in a rather short time, backed by very good data. During the same time, you're doing actual work that helps the organisations.

Please let me know if you're interested in chatting about all of this. I've learned a lot just since January when I launched the Altruistic Agency. I'd also be happy to share my application to the EA Infrastructure Fund that got me funding for the first six months.

Comment by Markus Amalthea Magnuson (peppersghost) on Announcing the Future Fund · 2022-03-09T15:33:49.677Z · EA · GW

Wonderful news. Do you have an idea of when the next open funding round will be? Or how often you will be open for applications, in general? I'm trying to determine how the upcoming March 21 deadline fits into my current plans for 2022.

Comment by Markus Amalthea Magnuson (peppersghost) on Common technical needs of EA organisations · 2022-02-18T12:29:25.602Z · EA · GW

The point of the post is to be mostly descriptive. I don't see where you get the idea I'm trying to push any particular tech. I have not advised anyone to replace their web platform unless they have explicitly asked for it, and I think WordPress is an excellent choice. Using website builders is and has always been, a trade-off between user convenience and developer convenience, and if you work in this industry you know this. I think you're reading far too much into what I'm reporting here. Also, I've worked as a full-stack web developer for 15+ years, at larger companies, startups, web agencies, and as a freelancer.

To clarify what I mean by system-level thinking:

Instead of thinking of every EA org as an isolated entity with isolated needs, I think of all of them as a collective. There is technical knowledge to be transferred between them and improvements that can be made to all organisations at once. For example, several organisations I've talked to have an interest in creating a job board, or improving the one they have. At the same time, I have also talked to 80,000 Hours about their plans for their job board and the fact they are hiring a director to run that specific project. This tells me there's an opportunity for them to expand the functionality of their job board to a point where several of the other organisations don't need one of their own. Maybe creating job boards for smaller organisations shouldn't be a priority until it's clear what's happening with this popular, already existent job board.

I just think it's a good idea to think about the needs of all EA organisations at once because there might also be solutions that allow meeting them for everyone while avoiding duplicating work. 

Comment by Markus Amalthea Magnuson (peppersghost) on Would free tech support and development be useful to your EA organisation? · 2021-10-09T17:36:41.903Z · EA · GW

Thank you, this is very extensive and informative.

Comment by Markus Amalthea Magnuson (peppersghost) on [EA Database] Update and calls for help · 2016-08-03T07:25:10.428Z · EA · GW

I think the second best option is to put this on its own website, something like or such. Then we would have complete freedom in modifying all parts (design, structure, meta data and similar for search engines, social link integration, etc. etc.) and it's more focused, easier to remember/point to/etc.

I did something similar with that is entirely based on the existing transhumanist FAQ that I felt was pure gold content-wise, but hidden deep within the H+ website, poorly laid out, not working at all on mobile, and many other things. This website is now the third result on Google when searching "what is transhumanism".

The website is hosted on GitHub meaning anyone with an account can suggest edits through pull requests, or make them directly if they have write access. Although decentralization can be a useful principle, I've found that having some sort of smaller group of editors makes vetting, structuring, and publishing a smoother process and yields a higher quality end result. Anyone is still free and encouraged to provide materials, of course.

Considering your point of view regarding difficulty-then-topic versus topic-then-difficulty, I think you are right and that your suggested structure is the better one. Topic is probably a more natural way of digging in for visitors of any level. As long as difficulty is clearly communicated (words, color coding, etc.) it will be fairly easy to navigate in the sense at the same time.

Maybe we could make an inventory of existing reading lists that could possibly be considered succeeded by this new one, and once up and running that the maintainers of these can be contacted about the possibility of being merged/redirected to the new website. (Obviously some existing ones should not follow this path and will instead be linked to from this website.) In other words, an inventory of potential "mergers" versus "co-existers".

These are some of my thoughts on your questions, looking forward to others' views.

Comment by Markus Amalthea Magnuson (peppersghost) on [EA Database] Update and calls for help · 2016-08-02T13:09:27.407Z · EA · GW

I have been thinking about this. One risk is that of increasing resource proliferation, i.e. that by adding more web pages for some purpose existing ones already fulfill, it will actually make it harder for someone to find either, or deem any of them more "authoritative" than others.

So one vast improvement to this project on top of the already excellent work you have done to extend the reading list/library and that everyone involved have done by providing the reading in the first place, would be to think about where it should live and how it is presented.

Judging by what happens when you google "effective altruism" the obvious place to keep a "semi-official" reading list/library is somewhere at

Specifically, I would suggest that clicking "Learn more" on the start page, then "Read more about EA" at the bottom of the page, should end up on another page on the same website with this entire reading list, divided into the three suggested levels (beginner etc.) so that it is obvious where to jump in. (The read more button currently points at which is a somewhat sparse and partly outdated list of reading.)

Which brings up another important aspect that could help this project a lot: to make sure outdated resources (or those that have been superseded by better ones for the same purposes) disappear from the web altogether, or even redirects to the new page once it exists. This obviously needs to happen through friendly discussion with those behind all the other pages. To make sure people end up at some pages rather than others, combined with other search engine optimization techniques, is what could really make a difference when it comes to steering people in the right direction.

However great the reading list (and let me add again that it is really great!) it won't matter if people don't find it, and the best way to make that happen is to put it in the most obvious and easiest to find place.

Sorry for the braindumpy quality of this reply, please let me know if anything is unclear and I'll try to clarify it. What are your general thoughts on the above?

P.S. By profession I'm a web developer and I'd love to help out with the more technical parts of this project, if that kind of help is needed.