Amber Dawn's Quick takes

Hi!

I haven't been very active on the Forum recently, so I thought I'd just pop in to remind people that I offer various editing-related services as a freelancer.

I'm based in London, but most of my work is remote. I offer copyediting, proofreading, substantive editing, conceptual support, review, and coaching. I've worked with many individuals and organisations in the EA and AI safety worlds: for example, I copyedited METR's recent Frontier Risk Report and the International AI Safety Report (2025 and 2026). 

If you'd like to work with me, book a short meeting or email me at [email protected].

Application forms for EA jobs often give an estimate for how long you should expect it to take; often these estimates are *wildly* too low ime. (And others I know have said this too). This is bad because it makes the estimates unhelpful for planning, and because it probably makes people feel bad about themselves, or worry that they're unusually slow, when they take longer than the estimate. 

Imo, if something involves any sort of writing from scratch, you should expect applicants to take at least an hour, and possibly more. (For context, I've seen application forms which say 'this application should take 10 minutes' and more commonly ones estimating 20 minutes or 30 minutes).

It doesn’t take long to type 300 words if you already know what you’re going to say and don’t particularly care about polish (I wrote this post in less than an hour probably).  But job application questions —even ‘basic’ ones like ‘why do you want this job?’ and ‘why would you be a good fit?’-- take more time. You may feel intuitively that you’d be a good fit for the job, but take a while to articulate why. You have to think about how your skills might help with the job, perhaps cross-referencing with the job description. And you have to express everything in appropriately-formal and clear language.

Job applications are also very high-stakes, and many people find them difficult or ‘ugh-y’, which means applicants are likely to take longer to do them than they “should”, due to being stuck or procrastinating. 

Maybe hirers put these time estimates because they don’t want applicants to spend too long on the first-stage form (for most of them, it won’t pay off, after all!) This respect for people’s time is laudable. But if someone really wants the job, they *will* feel motivated to put effort into the application form.

There’s a kind of coordination problem here too. Let's imagine there's an application for a job that I really want, and on the form it says 'this application should take you approximately 30 minutes'. If I knew that all the other applicants were going to set a timer for half an hour, write what came to mind, then send off the form without polishing it too much, I also might do that. But as far as I know, they are spending hours polishing their answers. I don’t want to incorrectly seem worse than other candidates and lose out on the job just because I took the time estimates more literally than other people!

‘Aren’t you just unusually slow and neurotic?’
-No; I’d guess that I write faster than average, and I’m really not perfectionist about job applications.

Suggestion: if you’re hiring, include a link at the end of the application form where people can anonymously report how long it actually took them.

You don't have to be an asshole just because you value honesty 

In Kirsten's recent EA Lifestyles advice column (NB, paywalled), an anonymous EA woman reported being bothered about men in the community whose "radical honesty" leads them to make inappropriate or hurtful comments:
 

For example: radical honesty/saying true things (great sometimes, not fun when men decide to be super honest about their sexual attraction or the exact amount they’re willing to account for women’s comfort until the costs just “aren’t justified.” This kind of openness is usually pointless: I can’t act on it, I didn’t want to know, and now I’m feeling hurt/wary).

 

An implication is that these guys may have viewed the discomfort of their women interlocutors as a (maybe regretful) cost of them upholding the important value of honesty. I've encountered similar attitudes elsewhere in EA - ie, people being kinda disagreeable/assholeish/mean under the excuse of 'just being honest'.

I want to say: I don't think a high commitment to honesty inevitably entails being disagreeable, acting unempathetically, or ruffling feathers. Why? Because I don't think it's dishonest not to say everything that springs to mind. If that were the case, I'd be continually narrating my internal monologue to my loved owns, and it would be very annoying for them, I'd imagine. 

If you're attracted to someone, and they ask "are you attracted to me?", and you say "no" - ok, that's dishonest. I don't think anyone should blame people for honestly answering a direct question. But if you're chatting with someone and you think "hmm, I'm really into them", and then you say that - I don't think honesty compels that choice, any more than it compels you to say "hold up, I just was thinking about whether I'd have soup or a burger for dinner".

I don't know much about the Radical Honesty movement, but from this article, it seems like they really prize just blurting out whatever you think. I do understand the urge to do this: I really value self-expression. For example, I'd struggle to be in a situation where I felt like I couldn't express my thoughts online and had to self-censor a lot. But I want to make the case that self-expression (how much of what comes to mind can you express vs being required to suppress) and honesty are somewhat orthogonal, and being maximally honest (ie, avoiding saying false things) doesn't require being maximally self-expressive. 


 

Let’s research some impactful interventions! Would you come to an intervention evaluation 101 learning-together event in London?

I want to run an event where we get together and do some quick-and-dirty intervention evaluation research, to learn more about how it works. I know nothing about this so we’ll be learning together!

Where: (central?) London
When: a mutually-agreed weekend day
What: I’ll come up with a structure loosely based on (some stages of?) the AIM/Charity Entrepreneurship research process. We’ll research and compare different interventions addressing the same broad problem or cause area. For example, we might start by quickly ranking or whittling down a long list in the morning, and then do some deeper dives in the afternoon.  We’ll alternate between doing independent research and discussing that research in pairs or small groups.

If you're interested in coming, please DM me: I’ll use a WhatsApp chat to coordinate. No need to firmly commit at this stage!

Why?
I hope to:

-Better understand how charity evaluators and incubators such as GiveWell and AIM form their recommendations, so I feel more empowered to engage with their research and can identify personal cruxes

-Learn how to assess interventions in areas that I think are promising, but that haven’t been discussed or researched extensively by EAs

-Just learn more about the world?

The event could also be useful for people who want to test their fit for EA research careers, though that’s not my own motivation.

What cause area?
We’d vote on a cause area beforehand. My vision here is something like ‘an area in global health and development that seems very important, but that hasn’t been discussed, or has been discussed relatively minimally, by EAs’.

What next?
If this goes well, we could hold these events regularly and/or collaborate and co-work on more substantive research projects.

Is there a point to this? 
AIM’s process takes ~1300 hours and is undertaken by skilled professional researchers; obviously we’re not going to produce recommendations of anywhere near similar quality. My motivation is to become personally better-informed and better-engaged with the nitty gritty of EA/being impactful in the world, rather than to reinvent the GiveWell wheel. 
That said, we’re stronger together: if 50 people worked on assessing a cause area together, they’d only have to spend 26 hours each to collectively equal the AIM process. 26 hours isn’t trivial (and nor is 50 people), but it’s not crazy implausible either. If collectives of EAs are putting AIM-level amounts of hours into intervention evaluation in their spare time, seems like a win?

"Death in a Shallow Pond": A new-ish book on the 'drowning child' thought experiment and EA

TIL about this book: Death in a Shallow Pond: A Philosopher, A Drowning Child, and Strangers in Need, published September 2025, by David Edmonds. I can't find it mentioned on the Forum but apologies if I've missed it. I haven't read it, but according to the blurb, it discusses 'the experiences and world events that led Singer to make his radical case and how it moved some young philosophers to establish the Effective Altruism movement, which tries to optimize philanthropy'.

I found out about the book from this Philosophy Bites podcast. Based on the podcast, I get the impression that Edmonds doesn't identify as an EA but is overall positive about the movement. 

Some stuff that frustrates me about the ‘dating within EA’ conversation

This post is related to ‘Consider not sleeping around within the community’, to the smattering of (thankfully heavily downvoted) posts unironically saying there should be less polyamory in EA, and to various conversations I’ve had about this, both in public and private. It’s less polished and more of an emotional snapshot. I feel extremely triggered/activated by this and I’m a bit wary that I’m earning myself a reputation as the unhinged-rants-about-polyamory woman,  or that I’m sort of arguing against something that isn’t substantively “there”.But I also think that emotions are information, and since these conversations are nominally about “making EA good/safe for women”, my perspective as a woman matters.

The meta-level

-We are all talking past each other. Some people are talking about power dynamic relationships. Some are talking about conflicts of interest. Some are talking about polyamory. Some are talking about casual sex or dating within EA. I've even seen one comment saying 'no-one should date anyone within EA'.  I'm likely part of the problem here, but yeah, this is aggravating. 

-I’m generally very wary of somewhat-vague admonishments addressed to a large group, with the assumption that the people who “need to hear” the admonishment will accurately self-select in and those to whom it doesn’t apply will accurately realise that and ignore it. Like, consider a feminist inveighing against vaguely how men are “trash” and/or need to “do better”. I’m pretty against this kind of rhetoric (unless it comes with a *hyper-specific* call to action or diagnosis of the bad behavior), because I think that this will cause anxiety for conscientious, neurotic, feminist men who wouldn’t hurt a fly (and sometimes queer women and NBs, if it’s relating to attraction to women), whereas abusive and/or misogynistic men are just not going to care. 

Similarly, I do not think men will correctly self-assess as socially clumsy or as having lots of power. Owen Cotton-Barratt’s statement is instructive here: he completely failed to see his own power. (Also, incidentally, if I understand right he was monogamously partnered and wasn't deliberately trying to hit on the women he made uncomfortable, so a 'don't sleep around in the community' norm wouldn't have helped, here). I think the advice 'avoid hitting on people if you're socially clumsy', if taken seriously, would lead to lots of kind, reasonable men neurotically never hitting on anybody - even in appropriate social contexts when those advances would be welcome - whereas boundary-pushers and predators won’t care.

This sort of thing is especially dangerous in an EA context, since EAs take moral injunctions very literally and very seriously. I think this is why I feel defensive about this.

-These conversations are supposed to be about “making EA better/safer for women” whereas (a) it’s not clear that most of the posts are even by women (some are anon, and lots of the comments are from men) and (b) as a woman who dates people in the community, this just feels deeply counter-productive and Not Helpful. It’s possible that there are norms that are good for women overall but not me specifically, but I think this is far from established and I’m still not crazy about being collateral damage.

The object level

-I do think that if I had taken some people’s views about dating within the community seriously, I wouldn’t have the relationships I do. I want to defend the attitudes and behaviours that led to me and my partners forming positive relationships with each other.

-I think this kind of critique implies a view of the world I disagree with. 
(I) it implies that a large part of the problems in EA come from social clumsiness, or maybe social clumsiness + power. I’m just more cynical about this: while I don’t want to minimize the harm done by 'off' comments and awkward advances, I’m more concerned about stuff like rape, assault, or ongoing abuse (in workplaces, homes or relationships). And there have been plenty of allegations of those things!

 I don’t subscribe to an overly black-and-white view of people where the are either bad villains or good well-meaning citizens, but I don’t think that you end up raping or abusing people through 'social clumsiness'.

(Ii) it implies that power is inevitable and relationships are not.  Like, one way to prevent the unsavory interaction of power + relationships is to dissuade relationships. Another is to try to distribute power more equitably and give more power to people who are newer to the community, lower within organisational hierarchies, and who are structurally disadvantaged by things like gender. Similarly, in situations where a relationship conflicts with a professional role, I'd strongly want to prioritize preserving the relationship over preserving the role, just because for most people romantic relationships are very important and meaningful, whereas work relationships are instrumental. 

I also think this kind of attitude takes responsibility and agency away from men? It assumes that drama and offence is just a necessary consequence of sexual interaction, rather than *something that can be mitigated* when people develop a feminist consciousness (and other progressive consciousnesses like anti-racism) and work on their empathy and social and emotional skills. The view ‘to solve gender problems we need to stop/limit sex’ seems both very pessimistic and kind of sexist against men. Rather than telling men not to date or even not to have casual sex, I’d rather tell men (and other genders! Other genders aren’t exempt from this!) to try to build the maturity to handle these encounters well, while empowering women so that they feel they can push back directly against minorly-inappropriate behaviour, and be supported in the case of suffering majorly inappropriate or harmful behaviour. 

 

Weakly against asking people to explain downvotes/disagree-votes (even politely)

Quite often it'll be clear that a post/comment is being downvoted/disagree-voted, and someone - either the OP or just a reader who likes the comment/post - writes that they're surprised at the disagreement/downvoting, and they'd be interested to know why people are disagreeing or why they don't like it. 

Most of the time these requests are very polite and non-demanding, but I'm still (weakly) against them, because I think they contribute to an expectation that if you downvote/disagree-vote, you have to be willing and able to 'defend' your choice to do that. But this is a very high bar - if I was forced to defend verbally all of my voting choices - and in language according to Forum norms, no less (not "I think the post is dumb, what do you want from me") -  I would almost never vote. If people wanted to explain why they disagreed or disliked the post, they probably would have already commented! 

It's also asymmetric - I've never seen someone say "I don't understand why this is getting upvoted". So asking people to explain downvotes/disagreevotes might lead to a dynamic where there's a mild disincentive to downvote/disagree and no comparable disincentive to upvote/agree, which means that controversial posts would appear to have more artificially more upvotes/agree-votes than they 'deserve'. 

 

Why doesn't EA focus on equity, human rights, and opposing discrimination (as cause areas)?

KJonEA asks:

'How focused do you think EA is on topics of race and gender equity/justice, human rights, and anti-discrimination? What do you think are factors that shape the community's focus?'

In response, I ended up writing a lot of words, so I thought it was worth editing them a bit and putting them in a shortform. I've also added some 'counterpoints' that weren't in the original comment. 

To lay my cards on the table: I'm a social progressive and leftist, and I think it would be cool if more EAs thought about equity, justice, human rights and discrimination - as cause areas to work in, rather than just within the EA community. (I'll call this cluster just 'equity' going forward). I also think it would be cool if left/progressive organisations had a more EA mindset sometimes. At the same time, as I hope my answers below show, I do think there are some good reasons that EAs don't prioritize equity, as well as some bad reasons. 

So, why don't EAs priority gender and racial equity, as cause areas? 

1. Other groups are already doing good work on equity (i.e. equity is less neglected)

The social justice/progressive movement has got feminism and anti-racism pretty well covered. On the other hand, the central EA causes - global health, AI safety, existential risk, animal welfare -are comparatively neglected by other groups. So it kinda makes sense for EAs to say 'we'll let these other movements keep doing their good work on these issues, and we'll focus on these other issues that not many people care about'.

Counter-point: are other groups using the most (cost)-effective methods to achieve their goals? EAs should, of course, be epistemically modest; but it seems that (e.g.) someone who was steeped in both EA and feminism, might have some great suggestions for how to improve gender equality and women's experiences, effectively.

2. Equity work isn't cost-effective

EAs care a lot about cost-effectiveness, ie how much demonstrable good impact you can get for your money. Maybe lots of equity/diversity work is, though important, not cost-effective: ie it's expensive, and the benefit is uncertain.

Counter-point: maybe EAs should try to work out how one might do equity work cost-effectively. ('Social justice' as such is seen as a western/rich world thing, but the countries where EA organisations often work also have equity problems, presumably).

3. Equity isn't an EA-shaped problem

EAs focus on what I think of as 'technical' solutions to social problems - i.e., 'fixes' that can be unilaterally performed by powerful actors such as nonprofits or governments or even singular wealthy individuals. I see many equity issues as cultural problems - that is, a nonprofit can't just swoop in and offer a wonk-ish fix; whole swathes of people have to be convinced to see the world differently and change their behaviour. Obviously, governments and NGOs do work on equity issues, but a big part of the "solution" to (e.g.) sexism is just "people, especially guys, learn basic feminist principles and stop being sexist to women". This is really important to work on, but it's not the style of solution that EAs tend to be drawn to.

4. EA is STEM-biased and equity is Humanities-biased 

Related: historically many EAs have been from STEM or analytic philosophy academic backgrounds (citation needed: is this in the survey?) These EAs are naturally more drawn to science-and-maths-y problems and solutions, like 'how to cure diseases and distribute those cures' or 'how to align artificial intelligence with human values', since these are the types of problems they've learnt how to solve. More 'humanities'-ish problems and solutions - like 'what are the interpersonal dynamics that make people's lives better or worse and how can we change culture in the positive direction?' - are out of the modal EA's comfort zone. 

5. EAs are privileged and underrate the need for equity

There are lots of people with social privilege in EA: it's majority white, straight, male, middle-class, etc. (Though getting more diverse on gender and race all the time, I think, and already more diverse on sexuality and stuff like mental health and neurodiversity than many social groups, I'd guess). You might predict that socially-privileged people would care less about equity issues than people who are directly impacted by those issues (though obviously this is not inevitable; many privileged people do care about equity issues that don't affect them).

6. EA is apolitical and equity is left-coded

Perhaps relatedly, EA is 'officially' apolitical, and equity/discrimination issues are more associated with left-wing or liberal politics. In fact, most EAs are liberal or politically left, but a decent amount are centrist, apolitical, libertarian or conservative. These EAs might not be interested in equity/discrimination issues; they might think that they're overrated, or they might dislike standard progressive approaches to equity (i.e. they might be "anti-woke"). This political minority might be vocal or central enough to sway the consensus.

Fund me to research interesting questions?
 

Here’s a list of questions/topics I’d be interested to research. If you’re also interested in one of these questions and would like to fund me to research it, get in touch: you can email at [email protected], DM me on the Forum, or book a chat. It’s a bit of a long shot, but you don’t get what you don’t ask for XD

I’m also keen to hear about relevant work that already exists. I haven’t done much work yet on any of these questions, so it’s possible there’s already a lot of research on them that I’m not aware of.

1. Why do people treat each other badly?

The world is pretty bad. Some of that badness isn’t caused by humans, for example disease and natural death. But lots of the badness comes from humans doing bad things to each other: abuse, war, and failing to save metaphorical drowning children.

Why is this?

Scott Alexander fans might say ‘Moloch’. Therapy people might say ‘trauma’. Evopsych people might say ‘it maximises fitness’. Other people, I’m sure, would say other things. Who is (most) right?

2. Antidote to the curse of knowledge

The ‘curse of knowledge’ is the phenomenon whereby, if you know about something, it’s really hard to explain it to others, because it’s hard for you to imagine not knowing what you know, and therefore what needs explanation. This is one of the fundamental difficulties of teaching and explanatory writing.

Relatedly, I think there is also a ‘curse of competence’: if you’re teaching or explaining a topic, chances are you have some natural aptitude for that topic or skill. If you’re a maths teacher, chances are you’re naturally good at math, and learnt it quite easily when you were young. This makes it harder for you to empathise with people who really struggle.

I think it would be cool to do some research into ‘systematic ways to bypass the curse of knowledge’. This could either be a technique that explain-ers and teachers could use themselves, or a technique for a teacher and student, or explainer and explain-ee, to use collaboratively. Such a technique might involve asking certain questions, developing a typology of ‘reasons people don’t understand a thing’, coming up with intuitive ways of ‘breaking things down’, etc.

(I expect there is some useful research and thought on this out there, so it might just be a question of collating/distilling it)

3. Could we make a society where everyone loved their work?

It seems like an awful lot of people don’t like their jobs, the thing they have to do for approximately 40 hours a week. This seems bad.

Charles Fourier was an early socialist/anarchist thinker. He had this (bonkers? genius?) idea that one could set up a happy society by forming people into units such that for every job that needed to be done, there would be enough people in the unit who were innately passionate about doing that job.
His idea was that you could drive production purely by exploiting people’s passions, so you wouldn’t need to force anyone to work with external incentives.

This seems… great if you could make it work?

I envisage that for this project, I’d start by reading Fourier’s writings and trying to extract the non-bonkers elements, and then move on to studying more prosaic ways that people have tried to improve working conditions, such as labour unions, workers cooperatives, even career coaching.

4. Surveys to work out global priorities

I’ve posted about this before. If we want to do the most good, it seems important to get a granular sense of what the majority of people in the world actually want and value the most. If the population of the world could vote on what I should do with my donations or career, would they want me to work on global health, or longtermist causes, or something else entirely?  

5. Getting ‘open borders’ into the Overton window and/or research into advocacy to decrease anti-immigrant sentiment

6. Ideas that changed people’s lives: substack/blog series

I want to interview people about specific ideas that changed their life, then write posts based on that.

What sort of ideas?
e.g.
-theories or facts about how the world works (e.g. historical, scientific, economic, personal?)
-relationship skills (e.g. non-violent communication, authentic relating, ??)
-therapeutic techniques (e.g. IFS, CBT, ACT, loving-kindness meditation, ??)
-political ideas (e.g. critical race theory, labour theory of value, classical liberalism, ??)
-philosophies (e.g. Stoicism, utilitarianism, Quakerism, ??)
-practical ideas (e.g. productivity or planning systems, skills, ??)

What do you mean ‘changed your life’?

-made you decide to do a certain sort of work, or advocacy
-changed your day-to-day habits
-improved your wellbeing or mental health
-improved your relationships

I’ve started this one already as a spare-time project, but if someone funded it, I could afford to spend more time on it.

My hope for this project is it will both spread lots of good ideas, and also help me understand ‘how people and ideas interact’, which might in turn help me understand how one could best spread good/helpful ideas, if one wanted to do that.

7. Anarchism: ???
More of a broad topic than a question. I’m drawn to anarchism but have a bunch of questions about it. There is loads of writing on anarchism, so this might be less of a research project, more of a distillation project; for example, producing an ‘Anarchism 101 for Dummies’ explainer, or coming up with and framing anarchism-inspired ideas that could, with skilled advocacy, spread and catch on (for obvious reasons, I'm thinking less of political advocacy and change, more of cultural change or movements).

How much funding do you want?
Say up to £25,000, but (much!) less is also fine? I’m open to lots of possibilities. You could fund me to work on one of these questions full-time, or part-time, for a few months or a year. Or you could say ‘I’ll pay you to do 3 hours of research on Question 2 and see how you get on’. Or you could do something in the middle. Happy to talk specifics in DMs.

Will you research [other thing] instead? 
It very much depends on what the other thing is and how well it fits my skills and interests, but feel free to ask.

What are your qualifications to do this?
I did most of a PhD in Classics. I ended up leaving the PhD before finishing*, but for many years I enjoyed the course and produced research regularly, and I got some good feedback on my research.

I think I’m good at the type of research that involves staring at difficult questions, sitting with confusion, working with concepts, understanding and synthesizing complex texts, and thinking by writing. Happy to send you my CV and/or talk in more detail about my credentials.

*this was because of a combination of poor mental health and becoming disenchanted with my topic and academia in general

What’s the impact?
I haven’t done a detailed impact analysis for any of these questions, but my intuition is that they are all difficult to solve/make progress on, but potentially highly impactful if you do make progress. The impact case for me working on these questions is not, imo, that they are likely have more impact than malaria prevention/AI safety/other central EA areas, but that they might be the highest impact thing for me personally to be doing.

Questions? Feel free to DM, email ([email protected]) or book a call. While I am shamelessly plugging myself, I am also doing writing and editing stuff: more details here. 

 

Since the season of Draft Amnesty is upon us, a bit of mild self-promotion: you can hire me to help you turn your unwritten thoughts and messy drafts into posts. 

For example:

-if some sections of your post are in your head but not yet on the page, I can help you draft them
-if you feel self-conscious about your draft, I can quickly review it and fix or flag the biggest issues
-if you feel ugh-y or uncertain about posting or finishing your post, or if you have anxieties about posting more generally, I can talk you through that
-I agree with the Forum team that it can be very valuable to share even unpolished and unfinished drafts, but if you'd like to publish a more polished post, I can help with editing and structuring

I have collaborated on a bunch of Forum posts.

Happy drafting!