This is a post about mental health and disposition in relation to the alignment problem. It compiles a number of resources that address how to maintain wellbeing and direction when confronted with existential risk.
Many people in this community have posted their emotional strategies for facing Doom after Eliezer Yudkowsky’s “Death With Dignity” generated so much conversation on the subject. This post intends to be more touchy-feely, dealing more directly with emotional landscapes than questions of timelines or probabilities of success.
The resources section would benefit from community additions. Please suggest any resources that you would like to see added to this post.
Please note that this document is not intended to replace professional medical or psychological help in any way. Many preexisting mental health conditions can be exacerbated by these conversations. If you are concerned that you may be experiencing a mental health crisis, please consult a professional.
Preface to the 2nd Edition
This post was released in April 2022 under the same title. This April 2023 update features new resources in every section, with a particular emphasis on the Alignment Positions and People Resources sections. Within each section, resources have been thematically categorized for easier access.
Following the large capabilities leaps in the past year, these resources seem more important than ever. If you have suggestions for improving this post, for making it more accessible, or for new resources to add, please leave a comment or reach out to either Chris Scammell or DivineMango.
We hope you are all well and that you find this update helpful.
Introduction
There is no right way to emotionally respond to the reality of approaching superintelligent AI, our collective responsibility to align it with our values, or the fact that we might not succeed. As transformative AI approaches, we must ensure that we have the tools and resources to be okay. Here, the valence of “be okay” is your decision. This question could be rephrased “how can I thrive despite the alignment problem,” “how can I cope with the alignment problem,” “how can I overcome my fear of the alignment problem,” etc. Everyone needs to find their own question and their own answer.
At its foundation “being okay” is the decision to continue to live facing reality and the alignment problem directly, with internal stability and rationality intact. And as a high ideal, we’re going for some degree of inviolability, of unconditional wellbeing, the kind of wellbeing that holds onto “okayness” even if the probability of solving alignment drops to 0. It can be difficult to stand in some place of positive mental health and stability while facing the alignment problem; but it is a gift if we can do that for ourselves, and a gift if we can share it with others.
Fortunately, we don’t have to do this alone. Many community members have found ways to make sense of themselves, their work, and their lives in relation to the alignment problem, and they have kindly made their reflections and advice public.
Resources
Several resources on this subject (along with summaries) are cataloged below. While there are a number of general mental health resources on LW, the EA Forum, and elsewhere that form a great baseline, this post aims to be more specific by focusing on mental health with respect to the alignment problem. Here, we feature a wide variety of ideas and practices in the hope that you may filter through them to create and discover the approach that works for you.
Human brains come in many shapes – we all have different internal subagent dynamics, motivational systems, values, needs, triggers for joy and fear, etc. Because of this variability, an approach that is great for one person may be bad for another. Some of you may need to take time to grieve. Some of you may need to focus on cultivating unconditional goodwill for yourself. Some of you may need to look squarely at existential terror and transmute it into motivation. As you read this article and browse these resources, remember to check in with yourself to see which approaches feel promising for you, given your past experience and your current mental landscape.
Alignment Positions
This section brings together posts on the subject of confronting despair of Doom on an emotional and practical level, categorized broadly by whether they focus on wellbeing or determination. These articles mostly focus on mental-emotional stances and philosophies, rather than actions.
Emotional Orientation & Wellbeing
Ruby:A Quick Guide to Confronting Doom. Start here. This post is exactly about this subject, and it is a good preface to reading the opinions below with the appropriate epistemic distance.
My guess is that people who are concluding P(Doom) is high will each need to figure out how to live with it for themselves. My caution is just that whatever strategy you figure out should keep you in touch with reality (or your best estimate of it), even if it’s uncomfortable.
Eliezer Yudkowsky: MIRI announces new ‘Death with Dignity’ strategy. Partially in jest, this post advocates that a good orientation for dealing with the alignment problem is to take actions that generate “dignity.” There has been debate in the community over both the aesthetics and content of the post, but there’s a coherent takeaway that I think most can agree on: try to be rational, and failing that, develop a good deontological strategy that protects against irresponsible action. There are a number of other Yudkowsky posts below which add nuance to this framework, and his Coming of Age sequence discusses alignment in Beyond the Reach of God.
So don’t get your heart set on that “not die at all” business. Don’t invest all your emotion in a reward you probably won’t get. Focus on dying with dignity—that is something you can actually obtain, even in this situation.
Valentine:Here’s the exit. In this controversial post, Valentine offers an escape from AGI terror for those who really need it. He claims that, in most cases, alignment discourse on LW is not about clear-headedly orienting to the (very real) problem of AI risk – it is about addictively obfuscating bodily-emotional pain with intense thought. His recommendation, for those who are done cooking their nervous systems, is to “land on earth and get sober.”
If your body’s emergency mobilization systems are running in response to an issue, but your survival doesn’t actually depend on actions on a timescale of minutes, then you are not perceiving reality accurately. Which is to say: If you’re freaked out but rushing around won’t solve the problem, then you’re living in a mental hallucination.
Duncan Sabien and Gretta Duleba: A Way to Be Okayand Another Way to Be Okay. Written collaboratively and in parallel, these posts touch directly on ways to be okay in the face of AI risk. The first post builds towards the Stoic-flavored idea of placing your self-evaluation in your own actions on alignment, rather than the outcome of alignment. The latter provides a mix of stances and strategies for being okay that orbit around ideas of presence and acceptance.
If you locate your identity in being the sort of person who does the best they can, given what they have and where they are, and if you define your victory condition as I did the best that I could throughout, given what I had and where I was, then while the tragedy of dying (yourself) or having the species/biosphere end is still really quite large and really quite traumatic, it nevertheless can’t quite cut at the core of you. – A Way to Be Okay
Tsvi: Please don’t throw your mind away. Often, Tsvi sees people sacrifice intellectual play and genuine intrinsic interest so that they can do more alignment-relevant things. He claims that doing this erases an important, bright part of you. Therefore, he strongly recommends against “throwing your mind away” in this fashion, and instead advocates that you let your mind play more.
It might also help to think of having fun sort of like walking: you know how in some sense, and you even have an instinct for it; having fun, if you’ve forgotten, is more a question of letting those circuits—which don’t require justification and just do what they do because that’s what they do—letting those circuits do what they do, and enjoying that those circuits do what they do. Basically the main thing here is just: there’s a thing called your mind, your mind likes to play seriously, and consider not preventing your mind from playing seriously.
Anna Salamon: What should you change in response to an “emergency”? And AI risk. This article discusses when “borrowing from the future” (e.g. working to exhaustion, neglecting personal enjoyment/needs) makes sense and when it doesn’t. Although AI is an emergency in an important sense, for most people it is not an emergency where burning out right now will better address the problem compared to working more sustainably over time.
Our best shot probably does mean paying attention to AI and ML advances, and directing some attention that way compared to what we’d do in a world where AI did not matter. It probably does mean doing the obvious work and the obvious alignment experiments where we know what those are, and where we can do this without burning out our long-term capacities. But it mostly doesn’t mean people burning themselves out, or depleting long-term resources in order to do this.
Justis: Maybe AI risk shouldn’t affect your life plan all that much. This post argues against letting doomerism seep into your decision-making, both because it is a strong meme (and thus likely overstated) and because predicting the future (even in the mid-term) is hard. Though its main argument depends on P(Near-term Doom) being somewhat small (< 0.4?), this post may be a helpful counterbalance for some people who are particularly doomy.
What does [a 2% chance of AI apocalypse] actually feel like?
- The odds of dying in a car crash over your lifetime areabout 1%. - The odds of dying of an opioid overdose, across the US population in general, are about1.5%. - The odds of dying of cancer are about14%.
So say you’re considering having a kid. It’s reasonable to worry a little that they’ll be killed by AI, perhaps even when they’re still young. Just like it’s reasonable to make sure they understand that it’s important to wear a seatbelt, and to get screened if they find any weird lumps when they’re older… And it may be correct that AI kills us all. But risk is just part of making life plans. We deal with low risks of horrifying outcomes all the time.
Zac Hatfield-Dodds: Concrete Reasons for Hope about AI. While many of the linked articles are about reactions to probable doom, this article offers a different view. According to Zac, alignment may not be as hopeless as it seems, even while remaining important to work on. Being optimistic about outcomes is okay. It is also okay to take the position that “everything is going to be okay” (alignment by default)!
While the situation is very scary indeed and often stressful, the x-risk mitigation community is a lovely and growing group of people, there’s a large frontier of work to be done, and I’m pretty confident that at least some of it will turn out to be helpful. So let’s get (back) to work!
Determination & Decisiveness
Jeffrey Ladish: Don’t die with dignity, instead play your outs. A response to Yudkowsky’s post, Ladish argues for an MTG-inspired strategy of “playing your outs,” or responding to a low-odds-of-success-future by looking ahead for what opportunities and affordances might still be available.
The framing doesn’t shy away from the fact that winning is unlikely. But the action is “playing” rather than “dying”. And the goal is “outs” rather than “dignity”. Again, I think the difference is in connotation and not actually strategy. To actually find outs, you have to search for solutions that might work, and stay focused on taking actions that improve our odds of success. When I imagine a Magic player playing to their outs, I imagine someone careful and engaged, not resigned. When I imagine someone dying with dignity, a terminally ill patient comes to mind. Peaceful, not panicking, but not fighting to survive.
TurnTrout: Emotionally Confronting a Probably Doomed World. A response to Yudkowsky’s post, TurnTrout argues that we should decouple our emotional response from the probability of doom and escape the idea that we are “living in a tragedy.” TurnTrout’s Swimming Upstream talks about earlier decisions to confront the alignment problem.
We do not live in a story. We can, in fact, just assess the situation, and then do what makes the most sense, what makes us strongest and happiest. The expected future of the universe is—by assumption—sad and horrible, and yet where is the ideal-agency theorem which says I must be downtrodden and glum about it?
Nate Soares: The Dark World. The fourth section of Soares’ Replacing Guilt series discusses a way to look squarely at the darkness in the world without crumbling under its weight.
So don’t let despair or hopelessness weigh you down. Instead, let them be a reminder: those are feelings you can only get from something worth saving. There are things here that are worth fighting for. If you begin to despair, then let that feeling be a reminder of what could be, and let everything that this world isn’t be your fuel. – Dark, Not Colorless
John Wentworth: We Choose To Align AI and The Plan. Together, these posts are a summary of Wentworth’s emotional position with respect to alignment and his specific plan to work on the problem. He offers a perspective that the magnitude of the challenge is reason for inspiration, not despair.
When people first seriously think about alignment, a majority freak out. Existential threats are terrifying… but for someone who wants the challenge, the emotional response is different.The problem is terrifying? Our current capabilities seem woefully inadequate? Good; this problem is worthy. The part of me which looks at a rickety ladder 30 feet down into a dark tunnel and says “let’s go!” wants this. The part of me which looks at a cliff face with no clear path up and cracks its knuckles wants this. The part of me which looks at a problem with no clear solution and smiles wants this. The response isn’t tears, it’s “let’s fucking do this”.
Richard Ngo: My Attitude Towards Death. Ngo’s post discusses fear of death, and his optimism for the future. He implies a strategy of “conversing” with his fear and trying to reassure it as a method for better integrating concerns.
Can I assure [the part of me that fears dying] that I’ll still try hard to avoid death if it becomes less scared? One source of assurance is if I’m very excited about a very long life—which I am, because the future could be amazing… Since I believe that we face significant existential risk this century, working to make humanity’s future go well overlaps heavily with working to make my own future go well. I think this broad argument has helped make the part of me that’s scared of death more quiescent.
Holden Karnofsky: Call to Vigilance. In the last post in Karnofsky’s The Most Important Century sequence, he describes that his emotional response to the alignment problem (and other challenges of our time) is one of intense, mixed emotions. But instead of telling people to rush in to “do something,” he advocates people should take robustly good actions, remain aware, and put themselves in positions to help in the future.
When confronting the “most important century” hypothesis, my attitude doesn’t match the familiar ones of “excitement and motion” or “fear and avoidance.” Instead, I feel an odd mix of intensity, urgency, confusion and hesitance. I’m looking at something bigger than I ever expected to confront, feeling underqualified and ignorant about what to do next. This is a hard mood to share and spread, but I’m trying.
In December 2022, at the Bay Area Secular Solstice, Clara Collier gave a poignant reading from C.S. Lewis about living and dying with dignity in the face of existential risk. It was a beautiful, harrowing moment, and, despite any disagreements you might have with it, the passage she read feels like a good emotional capstone for this section.
“In one way, we think a great deal too much of the atomic bomb. “How are we to live in an atomic age?” I am tempted to reply: “Why, as you would have lived in the sixteenth century when the plague visited London almost every year, or as you would have lived in a Viking age when raiders from Scandinavia might land and cut your throat any night; or indeed, as you are already living in an age of cancer, an age of syphilis, an age of paralysis, an age of air raids, an age of railway accidents, an age of motor accidents.”
In other words, do not let us begin by exaggerating the novelty of our situation. Believe me, dear sir or madam, you and all whom you love were already sentenced to death before the atomic bomb was invented: and quite a high percentage of us were going to die in unpleasant ways… It is perfectly ridiculous to go about whimpering and drawing long faces because the scientists have added one more chance of painful and premature death to a world which already bristled with such chances and in which death itself was not a chance at all, but a certainty.
… If we are all going to be destroyed by an atomic bomb, let that bomb when it comes find us doing sensible and human things: working, teaching, reading, listening to music, bathing the children, playing tennis, chatting to our friends over a pint and a game of darts—not huddled together like frightened sheep and thinking about bombs.”
– On Living in an Atomic Age
General Positions and Advice
These posts provide relevant opinions and guidance that are not directly about existential risks from AI.
On rising to the challenge: Yudkowsky and others: Challenging the Difficult and Heroic Responsibility. These hyperlinks go to collections of LW posts that advocate for building the internal drive to tackle problems as serious as alignment. These posts may be especially useful for individuals struggling with “helplessness” by transforming that feeling into action. Others may feel overburdened already and crumble under greater feelings of responsibility and pressure. Again: do what is best for you!
When you’re confused about a domain, problems in it will feel very intimidating and mysterious, and a query to your brain will produce a count of zero solutions. But you don’t know how much work will be left when the confusion clears. Dissolving the confusion may itself be a very difficult challenge, of course. But the word “impossible” should hardly be used in that connection. Confusion exists in the map, not in the territory. So if you spend a few years working on an impossible problem, and you manage to avoid or climb out of blind alleys, and your native ability is high enough to make progress, then, by golly, after a few years it may not seem so impossible after all. But if something seems impossible, you won’t try. - On Doing the Impossible
On how to overcome negative emotions:Replacing Guilt by Nate Soares. This foundational sequence has helped many people in the LW community transform feelings of guilt, resistance, sorrow, imposter syndrome, and other difficult emotions into inspiration. Despite the title, its scope is much larger than “guilt” and is a great starting place for any reader.
When all is said and done, Nature will not judge us by our actions; we will be measured only by what actually happens. Our goal, in the end, is to ensure that the timeless history of our universe is one that is filled with whatever it is we’re fighting for. For me, at least, this is the underlying driver that takes the place of guilt: Once we have learned our lessons from the past, there is no reason to wrack ourselves with guilt. All we need to do, in any given moment, is look upon the actions available to us, consider, and take whichever one seems most likely to lead to a future full of light.
On accepting sorrow and fear: Yudkowsky’s Feeling Rational and Luke Muehlhauser’s Musks’ Non-missing Mood. In contrast with some of the posts above that encourage decoupling emotions from probabilities of doom, these posts offer the perspective that negative emotions are not only a natural, but also a rational response. A related piece of advice from Hazard is to not confuse ignoring “useless” emotions for healthy emotional processing. For those confronting negative emotions who would like to accept and work with them, these posts may offer some insight.
When something terrible happens, I do not flee my sadness by searching for fake consolations and false silver linings. I visualize the past and future of humankind, the tens of billions of deaths over our history, the misery and fear, the search for answers, the trembling hands reaching upward out of so much blood, what we could become someday when we make the stars our cities, all that darkness and all that light—I know that I can never truly understand it, and I haven’t the words to say. Despite all my philosophy I am still embarrassed to confess strong emotions, and you’re probably uncomfortable hearing them. But I know, now, that it is rational to feel. – Feeling Rational
On working with imposter syndrome: Eight years ago, Luke Muehlhauser recommended If you’re an “AI safety lurker,” now would be a good time to de-lurk. But imposter syndrome and self-doubt can prevent people from raising their hand. Yudkowsky’s Hero Licensing talks about his own experience questioning the value of his work. Scott Alexander’s Parable of the Talents, Nicole Ross’ Desperation Hamster Wheels, and Luisa Rodriguez’s My experience with imposter syndrome are not specific to alignment, but they offer some advice on how to work with feelings of inadequacy. That said, self-worth is a deeper subject than just imposter syndrome and likely needs to be addressed outside of the context of productivity and alignment entirely.
When someone feels sad because they can’t be a great scientist, it is nice to be able to point out all of their intellectual strengths and tell them “Yes you can, if only you put your mind to it!” But this is often not true. At that point you have to say “f@#k it” and tell them to stop tying their self-worth to being a great scientist. And we had better establish that now, before transhumanists succeed in creating superintelligence and we all have to come to terms with our intellectual inferiority. – Parable of the Talents
On being honest about concerns: Katja Grace Beyond fire alarms: freeing the groupstuck. This post is primarily a response to Yudkowsky’s “There is No Fire Alarm for AGI,” but offers relevant ideas for how to deal with situations where one is afraid of looking silly for being overly-concerned about AI risk.
Practice voicing your somewhat embarrassing concerns, to make it easier for others to follow (and easier for you to do it again in future)… React to others’ concerns that don’t sound right to you with kindness and curiosity instead of laughter. Be especially nice about concerns about risks in particular, to counterbalance the special potential for shame there [or about people raising points that you think could possibly be embarrassing for them to raise]. – Beyond fire alarms
On overcoming avoidance: Anna Salamon’s Flinching away from the Truth and Making your explicit reasoning trustworthy. There are many other reasons people may not want to acknowledge the alignment problem, but one that isn’t addressed by the above points is avoidance due to the concern of mistaken beliefs and going down lines of thinking that lead to seductive but inaccurate conclusions. Anna’s posts may offer reassurance for those who are hesitant to engage fully with alignment ideas through their own reasoning, rather than relying on others’ positions.
“I don’t want to think about that! I might be left with mistaken beliefs!” tl;dr: Many of us hesitate to trust explicit reasoning because we haven’t built the skills that make such reasoning trustworthy. Some simple strategies can help. – Making your explicit reasoning trustworthy
On facing death: In addition to Ngo’s My Attitude Towards Death, there are a number of LW posts on death that may be useful to this conversation. Some such as Joe Carlsmith’s Thoughts on Being Mortal aren’t about alignment but confront fear of death directly, while some such as Yudkowsky’s The Meaning that Immortality Gives to Life touch on the singularity but are more about avoiding death. Avoidance of death is likely the crux of most fear and sorrow around the alignment problem, so from a purely mental-health related standpoint, it may be meaningful to try to separate emotional response to death from the alignment problem itself. Finding ways to confront death directly may afford a deep inviolability to existential fear.
Sometimes, on the comparatively rare occasions when I experience even-somewhat-intense sickness or pain, I think back to descriptions like this, and am brought more directly into the huge number of subjective worlds filled with relentless, inescapable pain. These glimpses often feel like a sudden shaking off of a certain kind of fuzziness; a clarifying of something central to what’s really going on in the world; and it also comes with fear of just how helpless we can become. – Thoughts on Being Mortal
HPMOR advice on facing existential risk: Yudkowsky’sHarry Potter and the Methods of Rationality is not about AI alignment (Harry deals mostly with local-to-planetary-scale rather than cosmological/hyperexistential threats), but the depicted emotions and mental strategies have direct analogue. The story contains deep explorations of the internal experience of facing seemingly impossible odds, the burden of heroic responsibility, difficult tradeoffs and the necessity of sacrifice, and the motivation for rational action and self-improvement. A non-exhaustive list of chapters that might be useful (and would be much more useful in context with the whole sequence):
Ch 39: death, motivation for transhumanism Ch 43-46: fear, death, motivation for transhumanism Ch 56-58: optimizing against improbable odds, despair Ch 63: the burden of responsibility, longing for a normal life Ch 75: heroic responsibility Ch 79-82: sacrifice Ch 88: fear of expressing panic, bystander apathy Ch 89: accepting/rejecting an unacceptable reality Ch 110: guilt, shame Ch 111-115: optimizing against improbable odds, despair Ch 117: guilt, sacrifice
In the large majority of cases, if you want to improve your mental health, you need to start doing something different in your life, rather than just think. Below are some tools and practices that span from interventions aimed at quickly cutting through negative states to longer-term practices aimed at building up more sustainable wellbeing.
Self-love: There are severalgreatresources that offer tools and frames for cultivating greater goodwill towards yourself. These practices may be particularly useful for working with anxiety, imposter syndrome, or fear. Thinking about AI x-risk can be very stressful, so make sure that you check in with how you’re doing and care for yourself.
Deliberate grief: For some, the most cathartic response to the possibility of alignment failure may be grief. Raemon grieves deliberately, which to him is a “process of noticing ‘Oh man, I sure do seem to be clinging to something that is maybe not real, or is gone now, or is no longer serving me. It’s starting to look like that clinging is shooting myself in foot, and it’d be nice if I could stop.’ And then… grieving, on purpose.” Valentine has also written about grief in The art of grieving well.
Focusing and Noticing: If you’re not sure what you “feel” about the alignment problem, or emotions you’ve previously felt are out of reach, the techniques of Focusing and Noticing can help. These methods bring awareness to sensations within the body, which increases clarity and affords an opportunity to do something about the feelings. For example, it may help uproot unconscious motivations that may be driving undesirable habits (procrastination, doomscrolling, etc.), or may help with anxiety or self-doubt.
Dark Arts: “Dark Arts” is a colloquial term for methods which involve deception or believing untrue things, such as intentional compartmentalization, inconsistency, or modifying terminal goals. These methods are not recommended for everyone, but they may help some individuals balance their life/productivity with intense feelings related to the alignment problem.
Productivity Sprints: Although this post is about mental health, not productivity, there are certain cases where feelings of despair and helplessness may be transformed by taking action. Nate Soare’s post here provides some practical advice, and Logan Riggs’ post demonstrates what it looks like to put that advice into practice.TurnTrout’s Problem relaxation as a tactic may also be helpful for those looking to get into alignment who find the scope of the problem too large.
The CFAR Handbook. If you would like to expand the mental toolkit you have for emotionally processing the alignment problem, the CFAR Handbook sequence might be useful for you. This sequence includes several frameworks for understanding yourself and your emotions.
People Resources
This section features therapists, coaches, and other entities who can provide support to those who may be struggling with their reactions to the alignment problem. Note that therapists should be consulted for more serious mental health struggles, rather than coaches. Prices given in parentheses indicate out-of-pocket per-session cost, where ranges indicate a sliding scale based on need.
There are many EA-adjacent therapists, but we are less sure which of them are familiar with alignment. If you know of any, please leave a comment with their name.
Therapists
Ewelina Tur (Poland, $120/£95/110€) specializes in providing therapy for effective altruists by drawing on her personal experience in the community (she co-founded EA Poland) and her wide knowledge of therapeutic modalities, including CBT and Mindfulness-Based Cognitive Therapy.
Damon Sasi (United States) also specializes in EA therapy, as well as therapy for rationalists. According to him, this includes “difficulties with motivation, guilt about impact in the world, desire for evidence-based methods of determining a course of action, therapy for non-monogamous relationships, and having a philosophy of grief and hardship that doesn’t include spirituality or religion.”
Igor Ivanov (United Kingdom, $110/£85/100€) is an EA therapist who specializes in CBT and Schema Therapy. He has several clients who work on AI safety, and he has written a post describing some alignment-related mental health issues he’s encountered in his practice, as well as a post about why impending AGI doesn’t make everything else unimportant.
Thomas Blank (Austria, $120/£95/110€) is more focused on CBT than other therapists listed here. He helps clients model their struggles by creating diagrams and flowcharts during sessions, which may be helpful for those who want a more concrete therapeutic workspace.
Daria Levin (United Kingdom, $100/£80/90€) has been working with EAs for many years and has a flexible therapeutic approach that integrates CBT, ACT, and Compassion-Focused Therapy, as well as many others. She helps clients with a wide range of struggles, from formal disorders such as depression, anxiety, and OCD, to less formal problems like perfectionism, burnout, and major life transitions.
Doris Schneeberger (Austria) specializes in psychoanalytically-oriented psychotherapy.
Markian Ozaruk (United States (NY), $150 with sliding scale) has several EA clients, many of whom chose him among several potential therapists. In addition to working with depression and anxiety (primarily through existential psychotherapy, along with IFS, CBT, and others as-needed), he also helps clients establish boundaries, work with their attachment style, and create a healthier locus of control while also cultivating agency.
Hannah Boettcher (United States (RI, MA, CT)) is a licensed clinical psychologist interested in upskilling to better help people in the AIS space. She has spoken on the 80K podcast about mental health challenges that come with trying to have a big impact.
Note: Although Hannah is fully booked at the time of posting (April 25, 2023), she is still interested in learning more about what people would want out of AIS-specialized therapy and the common failure modes that therapists fall into for them.
Coaches
Shay Gestal (United States, $125) is a health coach who partners with AI Safety Support to offer (previously free)weekly sessions to people who are interested in AI safety work.
Daniel Kestenholz (Denmark, $100/£80/90€) has completed over 900 coaching sessions, and he is familiar with AIS on a non-technical level. His clients include people at 80K, CEA, FHI, and Rethink Priorities.
Tee Barnett (Czechia, $250-500/£150-300/225€-450) is a “personal strategist” who has coached several EA leaders both personally and professionally. Coaches trained and supervised by Tee give sessions for a greatly reduced price ($50-100/£40-80/45€-90).
Terezie Kosíková (Czechia, $100/£80/90€) helped co-create an 8-week EA mental health program (workbook) and has coached several people through AI-induced existential dread.
Dave Cortright (United States) has a 25+ year tech background and is familiar with alignment. His coaching style has a Stoic emphasis, focusing on the client’s agency and what actions lie within their sphere of control without precluding compassion.
Kaj Sotala (Finland, $100-180/£80-140/90€-160) is a LW veteran and CFAR mentor with alignment research experience, and he is best known for his posts on Multiagent Models of Mind. He offers emotional coaching, primarily to clients who can meet during European daytime.
Sebastian Schmidt (United Kingdom, $300-400/£240-320/280€-370) provides coaching in six-session packages, where each session is 1.5 hours. According to one coachee, “He’s helped me clarify my long-term and short-term priorities and create systems to build new skills. These systems range from installing new habits to tracking systems to measure progress and hold myself accountable.”
Lynette Bye (United Kingdom, $200/£160/180€) is a productivity coach with over 2000 completed sessions. According to her 2020 impact evaluation, Lynette’s clients averaged an extra 25 productive hours per month due to her coaching.
Katie Glass (United Kingdom, free to unknown max) has coached internally at CEA and has experience working with people in both technical alignment and AI governance. She is well-equipped for helping with professional and personal development goals, but urges those with more serious mental health problems to seek professional help first.
alllinedup1234 (United States, free) is a therapist by training with a strong interest in alignment. Although he is only licensed in Ohio, he still wants to use his skills to help those who work in alignment. To that end, he is willing to provide through empathetic listening, brainstorming, coaching, and overall life skills free of charge.
Rickey Fukuzawa (Australia) is upskilling in AIS-centered coaching and will join Shay to give free sessions through AISS, provided that the funding comes through.
Ollie Bray (United Kingdom, donation) is an EA-adjacent meditation coach who gives hour-long sessions on a weekly to monthly basis.
Note: At the time of posting, Ollie is on indefinite leave. However, you can book a spot for coaching when he returns.
Other
EA Mental Health Navigator has a list of coaches and therapists who have experience working with effective altruists, though not necessarily with people in the alignment community. Nearly all of the above coaches and therapists were found in the Navigator and screened for their alignment familiarity.
The SSC Psychiat-list is a catalogue of SSC/rationalist recommended mental health providers that you can sort by price and strength of recommendation.
Rethink Wellbeing is an organization seeking to effectively improve EA mental health at scale. It has just completed the pilot round of its Effective Peer Support program, which finished with promising results.
AI Safety Support provides career advice to people interested in working on AI safety. Book a free call if you would like a fresh, informed perspective about your situation.
Sara Ness is the founder of Authentic Revolution and has offered to facilitate sessions for teams or groups working on AIS to talk through fears, needs, desires, and/or even concrete plans for navigating x-risk and their feelings about it.
A Final Note
Being happy and emotionally stable is instrumentally useful for making progress on alignment. But this post is written with the intention of increasing wellbeing, not productivity. We work on the alignment problem because we are driven by our deep care to protect the world we know, the one in which people experience joy and beauty and love. Wellbeing is instrumental for solving alignment, but more importantly, wellbeing is why we’re trying to solve it.
Mental Health and the Alignment Problem: A Compilation of Resources (updated April 2023)
This is a post about mental health and disposition in relation to the alignment problem. It compiles a number of resources that address how to maintain wellbeing and direction when confronted with existential risk.
Many people in this community have posted their emotional strategies for facing Doom after Eliezer Yudkowsky’s “Death With Dignity” generated so much conversation on the subject. This post intends to be more touchy-feely, dealing more directly with emotional landscapes than questions of timelines or probabilities of success.
The resources section would benefit from community additions. Please suggest any resources that you would like to see added to this post.
Please note that this document is not intended to replace professional medical or psychological help in any way. Many preexisting mental health conditions can be exacerbated by these conversations. If you are concerned that you may be experiencing a mental health crisis, please consult a professional.
Preface to the 2nd Edition
This post was released in April 2022 under the same title. This April 2023 update features new resources in every section, with a particular emphasis on the Alignment Positions and People Resources sections. Within each section, resources have been thematically categorized for easier access.
Following the large capabilities leaps in the past year, these resources seem more important than ever. If you have suggestions for improving this post, for making it more accessible, or for new resources to add, please leave a comment or reach out to either Chris Scammell or DivineMango.
We hope you are all well and that you find this update helpful.
Introduction
There is no right way to emotionally respond to the reality of approaching superintelligent AI, our collective responsibility to align it with our values, or the fact that we might not succeed. As transformative AI approaches, we must ensure that we have the tools and resources to be okay. Here, the valence of “be okay” is your decision. This question could be rephrased “how can I thrive despite the alignment problem,” “how can I cope with the alignment problem,” “how can I overcome my fear of the alignment problem,” etc. Everyone needs to find their own question and their own answer.
At its foundation “being okay” is the decision to continue to live facing reality and the alignment problem directly, with internal stability and rationality intact. And as a high ideal, we’re going for some degree of inviolability, of unconditional wellbeing, the kind of wellbeing that holds onto “okayness” even if the probability of solving alignment drops to 0. It can be difficult to stand in some place of positive mental health and stability while facing the alignment problem; but it is a gift if we can do that for ourselves, and a gift if we can share it with others.
Fortunately, we don’t have to do this alone. Many community members have found ways to make sense of themselves, their work, and their lives in relation to the alignment problem, and they have kindly made their reflections and advice public.
Resources
Several resources on this subject (along with summaries) are cataloged below. While there are a number of general mental health resources on LW, the EA Forum, and elsewhere that form a great baseline, this post aims to be more specific by focusing on mental health with respect to the alignment problem. Here, we feature a wide variety of ideas and practices in the hope that you may filter through them to create and discover the approach that works for you.
Human brains come in many shapes – we all have different internal subagent dynamics, motivational systems, values, needs, triggers for joy and fear, etc. Because of this variability, an approach that is great for one person may be bad for another. Some of you may need to take time to grieve. Some of you may need to focus on cultivating unconditional goodwill for yourself. Some of you may need to look squarely at existential terror and transmute it into motivation. As you read this article and browse these resources, remember to check in with yourself to see which approaches feel promising for you, given your past experience and your current mental landscape.
Alignment Positions
This section brings together posts on the subject of confronting despair of Doom on an emotional and practical level, categorized broadly by whether they focus on wellbeing or determination. These articles mostly focus on mental-emotional stances and philosophies, rather than actions.
Emotional Orientation & Wellbeing
Ruby: A Quick Guide to Confronting Doom. Start here. This post is exactly about this subject, and it is a good preface to reading the opinions below with the appropriate epistemic distance.
My guess is that people who are concluding P(Doom) is high will each need to figure out how to live with it for themselves. My caution is just that whatever strategy you figure out should keep you in touch with reality (or your best estimate of it), even if it’s uncomfortable.
Eliezer Yudkowsky: MIRI announces new ‘Death with Dignity’ strategy. Partially in jest, this post advocates that a good orientation for dealing with the alignment problem is to take actions that generate “dignity.” There has been debate in the community over both the aesthetics and content of the post, but there’s a coherent takeaway that I think most can agree on: try to be rational, and failing that, develop a good deontological strategy that protects against irresponsible action. There are a number of other Yudkowsky posts below which add nuance to this framework, and his Coming of Age sequence discusses alignment in Beyond the Reach of God.
So don’t get your heart set on that “not die at all” business. Don’t invest all your emotion in a reward you probably won’t get. Focus on dying with dignity—that is something you can actually obtain, even in this situation.
Valentine: Here’s the exit. In this controversial post, Valentine offers an escape from AGI terror for those who really need it. He claims that, in most cases, alignment discourse on LW is not about clear-headedly orienting to the (very real) problem of AI risk – it is about addictively obfuscating bodily-emotional pain with intense thought. His recommendation, for those who are done cooking their nervous systems, is to “land on earth and get sober.”
If your body’s emergency mobilization systems are running in response to an issue, but your survival doesn’t actually depend on actions on a timescale of minutes, then you are not perceiving reality accurately. Which is to say: If you’re freaked out but rushing around won’t solve the problem, then you’re living in a mental hallucination.
Duncan Sabien and Gretta Duleba: A Way to Be Okay and Another Way to Be Okay. Written collaboratively and in parallel, these posts touch directly on ways to be okay in the face of AI risk. The first post builds towards the Stoic-flavored idea of placing your self-evaluation in your own actions on alignment, rather than the outcome of alignment. The latter provides a mix of stances and strategies for being okay that orbit around ideas of presence and acceptance.
If you locate your identity in being the sort of person who does the best they can, given what they have and where they are, and if you define your victory condition as I did the best that I could throughout, given what I had and where I was, then while the tragedy of dying (yourself) or having the species/biosphere end is still really quite large and really quite traumatic, it nevertheless can’t quite cut at the core of you. – A Way to Be Okay
Tsvi: Please don’t throw your mind away. Often, Tsvi sees people sacrifice intellectual play and genuine intrinsic interest so that they can do more alignment-relevant things. He claims that doing this erases an important, bright part of you. Therefore, he strongly recommends against “throwing your mind away” in this fashion, and instead advocates that you let your mind play more.
It might also help to think of having fun sort of like walking: you know how in some sense, and you even have an instinct for it; having fun, if you’ve forgotten, is more a question of letting those circuits—which don’t require justification and just do what they do because that’s what they do—letting those circuits do what they do, and enjoying that those circuits do what they do. Basically the main thing here is just: there’s a thing called your mind, your mind likes to play seriously, and consider not preventing your mind from playing seriously.
Anna Salamon: What should you change in response to an “emergency”? And AI risk. This article discusses when “borrowing from the future” (e.g. working to exhaustion, neglecting personal enjoyment/needs) makes sense and when it doesn’t. Although AI is an emergency in an important sense, for most people it is not an emergency where burning out right now will better address the problem compared to working more sustainably over time.
Our best shot probably does mean paying attention to AI and ML advances, and directing some attention that way compared to what we’d do in a world where AI did not matter. It probably does mean doing the obvious work and the obvious alignment experiments where we know what those are, and where we can do this without burning out our long-term capacities. But it mostly doesn’t mean people burning themselves out, or depleting long-term resources in order to do this.
Justis: Maybe AI risk shouldn’t affect your life plan all that much. This post argues against letting doomerism seep into your decision-making, both because it is a strong meme (and thus likely overstated) and because predicting the future (even in the mid-term) is hard. Though its main argument depends on P(Near-term Doom) being somewhat small (< 0.4?), this post may be a helpful counterbalance for some people who are particularly doomy.
What does [a 2% chance of AI apocalypse] actually feel like?
- The odds of dying in a car crash over your lifetime are about 1%.
- The odds of dying of an opioid overdose, across the US population in general, are about 1.5%.
- The odds of dying of cancer are about 14%.
So say you’re considering having a kid. It’s reasonable to worry a little that they’ll be killed by AI, perhaps even when they’re still young. Just like it’s reasonable to make sure they understand that it’s important to wear a seatbelt, and to get screened if they find any weird lumps when they’re older… And it may be correct that AI kills us all. But risk is just part of making life plans. We deal with low risks of horrifying outcomes all the time.
Zac Hatfield-Dodds: Concrete Reasons for Hope about AI. While many of the linked articles are about reactions to probable doom, this article offers a different view. According to Zac, alignment may not be as hopeless as it seems, even while remaining important to work on. Being optimistic about outcomes is okay. It is also okay to take the position that “everything is going to be okay” (alignment by default)!
While the situation is very scary indeed and often stressful, the x-risk mitigation community is a lovely and growing group of people, there’s a large frontier of work to be done, and I’m pretty confident that at least some of it will turn out to be helpful. So let’s get (back) to work!
Determination & Decisiveness
Jeffrey Ladish: Don’t die with dignity, instead play your outs. A response to Yudkowsky’s post, Ladish argues for an MTG-inspired strategy of “playing your outs,” or responding to a low-odds-of-success-future by looking ahead for what opportunities and affordances might still be available.
The framing doesn’t shy away from the fact that winning is unlikely. But the action is “playing” rather than “dying”. And the goal is “outs” rather than “dignity”. Again, I think the difference is in connotation and not actually strategy. To actually find outs, you have to search for solutions that might work, and stay focused on taking actions that improve our odds of success. When I imagine a Magic player playing to their outs, I imagine someone careful and engaged, not resigned. When I imagine someone dying with dignity, a terminally ill patient comes to mind. Peaceful, not panicking, but not fighting to survive.
TurnTrout: Emotionally Confronting a Probably Doomed World. A response to Yudkowsky’s post, TurnTrout argues that we should decouple our emotional response from the probability of doom and escape the idea that we are “living in a tragedy.” TurnTrout’s Swimming Upstream talks about earlier decisions to confront the alignment problem.
We do not live in a story. We can, in fact, just assess the situation, and then do what makes the most sense, what makes us strongest and happiest. The expected future of the universe is—by assumption—sad and horrible, and yet where is the ideal-agency theorem which says I must be downtrodden and glum about it?
Nate Soares: The Dark World. The fourth section of Soares’ Replacing Guilt series discusses a way to look squarely at the darkness in the world without crumbling under its weight.
So don’t let despair or hopelessness weigh you down. Instead, let them be a reminder: those are feelings you can only get from something worth saving. There are things here that are worth fighting for. If you begin to despair, then let that feeling be a reminder of what could be, and let everything that this world isn’t be your fuel. – Dark, Not Colorless
John Wentworth: We Choose To Align AI and The Plan. Together, these posts are a summary of Wentworth’s emotional position with respect to alignment and his specific plan to work on the problem. He offers a perspective that the magnitude of the challenge is reason for inspiration, not despair.
When people first seriously think about alignment, a majority freak out. Existential threats are terrifying… but for someone who wants the challenge, the emotional response is different.The problem is terrifying? Our current capabilities seem woefully inadequate? Good; this problem is worthy. The part of me which looks at a rickety ladder 30 feet down into a dark tunnel and says “let’s go!” wants this. The part of me which looks at a cliff face with no clear path up and cracks its knuckles wants this. The part of me which looks at a problem with no clear solution and smiles wants this. The response isn’t tears, it’s “let’s fucking do this”.
Richard Ngo: My Attitude Towards Death. Ngo’s post discusses fear of death, and his optimism for the future. He implies a strategy of “conversing” with his fear and trying to reassure it as a method for better integrating concerns.
Can I assure [the part of me that fears dying] that I’ll still try hard to avoid death if it becomes less scared? One source of assurance is if I’m very excited about a very long life—which I am, because the future could be amazing… Since I believe that we face significant existential risk this century, working to make humanity’s future go well overlaps heavily with working to make my own future go well. I think this broad argument has helped make the part of me that’s scared of death more quiescent.
Holden Karnofsky: Call to Vigilance. In the last post in Karnofsky’s The Most Important Century sequence, he describes that his emotional response to the alignment problem (and other challenges of our time) is one of intense, mixed emotions. But instead of telling people to rush in to “do something,” he advocates people should take robustly good actions, remain aware, and put themselves in positions to help in the future.
When confronting the “most important century” hypothesis, my attitude doesn’t match the familiar ones of “excitement and motion” or “fear and avoidance.” Instead, I feel an odd mix of intensity, urgency, confusion and hesitance. I’m looking at something bigger than I ever expected to confront, feeling underqualified and ignorant about what to do next. This is a hard mood to share and spread, but I’m trying.
In December 2022, at the Bay Area Secular Solstice, Clara Collier gave a poignant reading from C.S. Lewis about living and dying with dignity in the face of existential risk. It was a beautiful, harrowing moment, and, despite any disagreements you might have with it, the passage she read feels like a good emotional capstone for this section.
“In one way, we think a great deal too much of the atomic bomb. “How are we to live in an atomic age?” I am tempted to reply: “Why, as you would have lived in the sixteenth century when the plague visited London almost every year, or as you would have lived in a Viking age when raiders from Scandinavia might land and cut your throat any night; or indeed, as you are already living in an age of cancer, an age of syphilis, an age of paralysis, an age of air raids, an age of railway accidents, an age of motor accidents.”
In other words, do not let us begin by exaggerating the novelty of our situation. Believe me, dear sir or madam, you and all whom you love were already sentenced to death before the atomic bomb was invented: and quite a high percentage of us were going to die in unpleasant ways… It is perfectly ridiculous to go about whimpering and drawing long faces because the scientists have added one more chance of painful and premature death to a world which already bristled with such chances and in which death itself was not a chance at all, but a certainty.
… If we are all going to be destroyed by an atomic bomb, let that bomb when it comes find us doing sensible and human things: working, teaching, reading, listening to music, bathing the children, playing tennis, chatting to our friends over a pint and a game of darts—not huddled together like frightened sheep and thinking about bombs.”
– On Living in an Atomic Age
General Positions and Advice
These posts provide relevant opinions and guidance that are not directly about existential risks from AI.
On rising to the challenge: Yudkowsky and others: Challenging the Difficult and Heroic Responsibility. These hyperlinks go to collections of LW posts that advocate for building the internal drive to tackle problems as serious as alignment. These posts may be especially useful for individuals struggling with “helplessness” by transforming that feeling into action. Others may feel overburdened already and crumble under greater feelings of responsibility and pressure. Again: do what is best for you!
When you’re confused about a domain, problems in it will feel very intimidating and mysterious, and a query to your brain will produce a count of zero solutions. But you don’t know how much work will be left when the confusion clears. Dissolving the confusion may itself be a very difficult challenge, of course. But the word “impossible” should hardly be used in that connection. Confusion exists in the map, not in the territory. So if you spend a few years working on an impossible problem, and you manage to avoid or climb out of blind alleys, and your native ability is high enough to make progress, then, by golly, after a few years it may not seem so impossible after all. But if something seems impossible, you won’t try.
- On Doing the Impossible
If you’re motivated to do something about alignment, there are many pragmatic posts on LW as well as non-LW resources like AI Safety Support, the AGI Safety Fundamentals Course, and 80,000 Hours.
On how to overcome negative emotions: Replacing Guilt by Nate Soares. This foundational sequence has helped many people in the LW community transform feelings of guilt, resistance, sorrow, imposter syndrome, and other difficult emotions into inspiration. Despite the title, its scope is much larger than “guilt” and is a great starting place for any reader.
When all is said and done, Nature will not judge us by our actions; we will be measured only by what actually happens. Our goal, in the end, is to ensure that the timeless history of our universe is one that is filled with whatever it is we’re fighting for. For me, at least, this is the underlying driver that takes the place of guilt: Once we have learned our lessons from the past, there is no reason to wrack ourselves with guilt. All we need to do, in any given moment, is look upon the actions available to us, consider, and take whichever one seems most likely to lead to a future full of light.
On accepting sorrow and fear: Yudkowsky’s Feeling Rational and Luke Muehlhauser’s Musks’ Non-missing Mood. In contrast with some of the posts above that encourage decoupling emotions from probabilities of doom, these posts offer the perspective that negative emotions are not only a natural, but also a rational response. A related piece of advice from Hazard is to not confuse ignoring “useless” emotions for healthy emotional processing. For those confronting negative emotions who would like to accept and work with them, these posts may offer some insight.
When something terrible happens, I do not flee my sadness by searching for fake consolations and false silver linings. I visualize the past and future of humankind, the tens of billions of deaths over our history, the misery and fear, the search for answers, the trembling hands reaching upward out of so much blood, what we could become someday when we make the stars our cities, all that darkness and all that light—I know that I can never truly understand it, and I haven’t the words to say. Despite all my philosophy I am still embarrassed to confess strong emotions, and you’re probably uncomfortable hearing them. But I know, now, that it is rational to feel. – Feeling Rational
On working with imposter syndrome: Eight years ago, Luke Muehlhauser recommended If you’re an “AI safety lurker,” now would be a good time to de-lurk. But imposter syndrome and self-doubt can prevent people from raising their hand. Yudkowsky’s Hero Licensing talks about his own experience questioning the value of his work. Scott Alexander’s Parable of the Talents, Nicole Ross’ Desperation Hamster Wheels, and Luisa Rodriguez’s My experience with imposter syndrome are not specific to alignment, but they offer some advice on how to work with feelings of inadequacy. That said, self-worth is a deeper subject than just imposter syndrome and likely needs to be addressed outside of the context of productivity and alignment entirely.
When someone feels sad because they can’t be a great scientist, it is nice to be able to point out all of their intellectual strengths and tell them “Yes you can, if only you put your mind to it!” But this is often not true. At that point you have to say “f@#k it” and tell them to stop tying their self-worth to being a great scientist. And we had better establish that now, before transhumanists succeed in creating superintelligence and we all have to come to terms with our intellectual inferiority. – Parable of the Talents
On being honest about concerns: Katja Grace Beyond fire alarms: freeing the groupstuck. This post is primarily a response to Yudkowsky’s “There is No Fire Alarm for AGI,” but offers relevant ideas for how to deal with situations where one is afraid of looking silly for being overly-concerned about AI risk.
Practice voicing your somewhat embarrassing concerns, to make it easier for others to follow (and easier for you to do it again in future)… React to others’ concerns that don’t sound right to you with kindness and curiosity instead of laughter. Be especially nice about concerns about risks in particular, to counterbalance the special potential for shame there [or about people raising points that you think could possibly be embarrassing for them to raise]. – Beyond fire alarms
On overcoming avoidance: Anna Salamon’s Flinching away from the Truth and Making your explicit reasoning trustworthy. There are many other reasons people may not want to acknowledge the alignment problem, but one that isn’t addressed by the above points is avoidance due to the concern of mistaken beliefs and going down lines of thinking that lead to seductive but inaccurate conclusions. Anna’s posts may offer reassurance for those who are hesitant to engage fully with alignment ideas through their own reasoning, rather than relying on others’ positions.
“I don’t want to think about that! I might be left with mistaken beliefs!” tl;dr: Many of us hesitate to trust explicit reasoning because we haven’t built the skills that make such reasoning trustworthy. Some simple strategies can help. – Making your explicit reasoning trustworthy
On facing death: In addition to Ngo’s My Attitude Towards Death, there are a number of LW posts on death that may be useful to this conversation. Some such as Joe Carlsmith’s Thoughts on Being Mortal aren’t about alignment but confront fear of death directly, while some such as Yudkowsky’s The Meaning that Immortality Gives to Life touch on the singularity but are more about avoiding death. Avoidance of death is likely the crux of most fear and sorrow around the alignment problem, so from a purely mental-health related standpoint, it may be meaningful to try to separate emotional response to death from the alignment problem itself. Finding ways to confront death directly may afford a deep inviolability to existential fear.
Sometimes, on the comparatively rare occasions when I experience even-somewhat-intense sickness or pain, I think back to descriptions like this, and am brought more directly into the huge number of subjective worlds filled with relentless, inescapable pain. These glimpses often feel like a sudden shaking off of a certain kind of fuzziness; a clarifying of something central to what’s really going on in the world; and it also comes with fear of just how helpless we can become. – Thoughts on Being Mortal
HPMOR advice on facing existential risk: Yudkowsky’s Harry Potter and the Methods of Rationality is not about AI alignment (Harry deals mostly with local-to-planetary-scale rather than cosmological/hyperexistential threats), but the depicted emotions and mental strategies have direct analogue. The story contains deep explorations of the internal experience of facing seemingly impossible odds, the burden of heroic responsibility, difficult tradeoffs and the necessity of sacrifice, and the motivation for rational action and self-improvement. A non-exhaustive list of chapters that might be useful (and would be much more useful in context with the whole sequence):
Ch 39: death, motivation for transhumanism
Ch 43-46: fear, death, motivation for transhumanism
Ch 56-58: optimizing against improbable odds, despair
Ch 63: the burden of responsibility, longing for a normal life
Ch 75: heroic responsibility
Ch 79-82: sacrifice
Ch 88: fear of expressing panic, bystander apathy
Ch 89: accepting/rejecting an unacceptable reality
Ch 110: guilt, shame
Ch 111-115: optimizing against improbable odds, despair
Ch 117: guilt, sacrifice
EA resources on general wellbeing and burnout. While not specific to alignment, it would be a mistake not to mention the wealth of information on the EA forum related to mental health such as Miranda Zhang’s Mental Health Resources tailored for EAs and Ewelina Tur’s List of Mental Health Resources. The EA forum also has a bunch of specific posts on burnout, like Elizabeth’s Burnout, Tessa’s Aiming for the minimum of self-care is dangerous, Julia Wise’s Cheerfully, and Logan Strohl’s My Model of EA Burnout.
Tools and Practices
In the large majority of cases, if you want to improve your mental health, you need to start doing something different in your life, rather than just think. Below are some tools and practices that span from interventions aimed at quickly cutting through negative states to longer-term practices aimed at building up more sustainable wellbeing.
Therapy: If you have significant mental health struggles, start by finding a therapist that works for you. The final section of this post features a list of alignment-familiar therapists and life coaches. Remember that therapy is not merely emotional support and has different goals from coaching.
Meditation: Kaj Sotala’s My attempt to explain Looking, insight meditation, and enlightenment in non-mysterious terms. This post explains how meditation practices can help people develop unconditional groundedness, even in the face of existential risk. Sotala’s sequence goes deeper in these ideas, and he also has a number of practical individual posts like Overcoming suffering: emotional acceptance.
EA-adjacent meditation coach Ollie Bray has written about making systematic progress in meditation through the cultivation of joy.
Self-love: There are several great resources that offer tools and frames for cultivating greater goodwill towards yourself. These practices may be particularly useful for working with anxiety, imposter syndrome, or fear. Thinking about AI x-risk can be very stressful, so make sure that you check in with how you’re doing and care for yourself.
Deliberate grief: For some, the most cathartic response to the possibility of alignment failure may be grief. Raemon grieves deliberately, which to him is a “process of noticing ‘Oh man, I sure do seem to be clinging to something that is maybe not real, or is gone now, or is no longer serving me. It’s starting to look like that clinging is shooting myself in foot, and it’d be nice if I could stop.’ And then… grieving, on purpose.” Valentine has also written about grief in The art of grieving well.
Focusing and Noticing: If you’re not sure what you “feel” about the alignment problem, or emotions you’ve previously felt are out of reach, the techniques of Focusing and Noticing can help. These methods bring awareness to sensations within the body, which increases clarity and affords an opportunity to do something about the feelings. For example, it may help uproot unconscious motivations that may be driving undesirable habits (procrastination, doomscrolling, etc.), or may help with anxiety or self-doubt.
Dark Arts: “Dark Arts” is a colloquial term for methods which involve deception or believing untrue things, such as intentional compartmentalization, inconsistency, or modifying terminal goals. These methods are not recommended for everyone, but they may help some individuals balance their life/productivity with intense feelings related to the alignment problem.
Productivity Sprints: Although this post is about mental health, not productivity, there are certain cases where feelings of despair and helplessness may be transformed by taking action. Nate Soare’s post here provides some practical advice, and Logan Riggs’ post demonstrates what it looks like to put that advice into practice. TurnTrout’s Problem relaxation as a tactic may also be helpful for those looking to get into alignment who find the scope of the problem too large.
The CFAR Handbook. If you would like to expand the mental toolkit you have for emotionally processing the alignment problem, the CFAR Handbook sequence might be useful for you. This sequence includes several frameworks for understanding yourself and your emotions.
People Resources
This section features therapists, coaches, and other entities who can provide support to those who may be struggling with their reactions to the alignment problem. Note that therapists should be consulted for more serious mental health struggles, rather than coaches. Prices given in parentheses indicate out-of-pocket per-session cost, where ranges indicate a sliding scale based on need.
There are many EA-adjacent therapists, but we are less sure which of them are familiar with alignment. If you know of any, please leave a comment with their name.
Therapists
Ewelina Tur (Poland, $120/£95/110€) specializes in providing therapy for effective altruists by drawing on her personal experience in the community (she co-founded EA Poland) and her wide knowledge of therapeutic modalities, including CBT and Mindfulness-Based Cognitive Therapy.
Damon Sasi (United States) also specializes in EA therapy, as well as therapy for rationalists. According to him, this includes “difficulties with motivation, guilt about impact in the world, desire for evidence-based methods of determining a course of action, therapy for non-monogamous relationships, and having a philosophy of grief and hardship that doesn’t include spirituality or religion.”
Igor Ivanov (United Kingdom, $110/£85/100€) is an EA therapist who specializes in CBT and Schema Therapy. He has several clients who work on AI safety, and he has written a post describing some alignment-related mental health issues he’s encountered in his practice, as well as a post about why impending AGI doesn’t make everything else unimportant.
Thomas Blank (Austria, $120/£95/110€) is more focused on CBT than other therapists listed here. He helps clients model their struggles by creating diagrams and flowcharts during sessions, which may be helpful for those who want a more concrete therapeutic workspace.
Daria Levin (United Kingdom, $100/£80/90€) has been working with EAs for many years and has a flexible therapeutic approach that integrates CBT, ACT, and Compassion-Focused Therapy, as well as many others. She helps clients with a wide range of struggles, from formal disorders such as depression, anxiety, and OCD, to less formal problems like perfectionism, burnout, and major life transitions.
Doris Schneeberger (Austria) specializes in psychoanalytically-oriented psychotherapy.
Markian Ozaruk (United States (NY), $150 with sliding scale) has several EA clients, many of whom chose him among several potential therapists. In addition to working with depression and anxiety (primarily through existential psychotherapy, along with IFS, CBT, and others as-needed), he also helps clients establish boundaries, work with their attachment style, and create a healthier locus of control while also cultivating agency.
Hannah Boettcher (United States (RI, MA, CT)) is a licensed clinical psychologist interested in upskilling to better help people in the AIS space. She has spoken on the 80K podcast about mental health challenges that come with trying to have a big impact.
Note: Although Hannah is fully booked at the time of posting (April 25, 2023), she is still interested in learning more about what people would want out of AIS-specialized therapy and the common failure modes that therapists fall into for them.
Coaches
Shay Gestal (United States, $125) is a health coach who partners with AI Safety Support to offer (previously free) weekly sessions to people who are interested in AI safety work.
Daniel Kestenholz (Denmark, $100/£80/90€) has completed over 900 coaching sessions, and he is familiar with AIS on a non-technical level. His clients include people at 80K, CEA, FHI, and Rethink Priorities.
Tee Barnett (Czechia, $250-500/£150-300/225€-450) is a “personal strategist” who has coached several EA leaders both personally and professionally. Coaches trained and supervised by Tee give sessions for a greatly reduced price ($50-100/£40-80/45€-90).
Terezie Kosíková (Czechia, $100/£80/90€) helped co-create an 8-week EA mental health program (workbook) and has coached several people through AI-induced existential dread.
Dave Cortright (United States) has a 25+ year tech background and is familiar with alignment. His coaching style has a Stoic emphasis, focusing on the client’s agency and what actions lie within their sphere of control without precluding compassion.
Kaj Sotala (Finland, $100-180/£80-140/90€-160) is a LW veteran and CFAR mentor with alignment research experience, and he is best known for his posts on Multiagent Models of Mind. He offers emotional coaching, primarily to clients who can meet during European daytime.
Sebastian Schmidt (United Kingdom, $300-400/£240-320/280€-370) provides coaching in six-session packages, where each session is 1.5 hours. According to one coachee, “He’s helped me clarify my long-term and short-term priorities and create systems to build new skills. These systems range from installing new habits to tracking systems to measure progress and hold myself accountable.”
Lynette Bye (United Kingdom, $200/£160/180€) is a productivity coach with over 2000 completed sessions. According to her 2020 impact evaluation, Lynette’s clients averaged an extra 25 productive hours per month due to her coaching.
Katie Glass (United Kingdom, free to unknown max) has coached internally at CEA and has experience working with people in both technical alignment and AI governance. She is well-equipped for helping with professional and personal development goals, but urges those with more serious mental health problems to seek professional help first.
alllinedup1234 (United States, free) is a therapist by training with a strong interest in alignment. Although he is only licensed in Ohio, he still wants to use his skills to help those who work in alignment. To that end, he is willing to provide through empathetic listening, brainstorming, coaching, and overall life skills free of charge.
Rickey Fukuzawa (Australia) is upskilling in AIS-centered coaching and will join Shay to give free sessions through AISS, provided that the funding comes through.
Ollie Bray (United Kingdom, donation) is an EA-adjacent meditation coach who gives hour-long sessions on a weekly to monthly basis.
Note: At the time of posting, Ollie is on indefinite leave. However, you can book a spot for coaching when he returns.
Other
EA Mental Health Navigator has a list of coaches and therapists who have experience working with effective altruists, though not necessarily with people in the alignment community. Nearly all of the above coaches and therapists were found in the Navigator and screened for their alignment familiarity.
The SSC Psychiat-list is a catalogue of SSC/rationalist recommended mental health providers that you can sort by price and strength of recommendation.
Rethink Wellbeing is an organization seeking to effectively improve EA mental health at scale. It has just completed the pilot round of its Effective Peer Support program, which finished with promising results.
AI Safety Support provides career advice to people interested in working on AI safety. Book a free call if you would like a fresh, informed perspective about your situation.
Sara Ness is the founder of Authentic Revolution and has offered to facilitate sessions for teams or groups working on AIS to talk through fears, needs, desires, and/or even concrete plans for navigating x-risk and their feelings about it.
A Final Note
Being happy and emotionally stable is instrumentally useful for making progress on alignment. But this post is written with the intention of increasing wellbeing, not productivity. We work on the alignment problem because we are driven by our deep care to protect the world we know, the one in which people experience joy and beauty and love. Wellbeing is instrumental for solving alignment, but more importantly, wellbeing is why we’re trying to solve it.