I spent 12 hours writing an article for LessWrong, calling for a more ‘hippie’ rationalism. Then I didn’t post it. This is why (plus a link to the article, explicitly not located on LessWrong) — LessWrong
I (and many others in the community) wish to see the universe unfold to its highest potential in the metaphorical palm of an AI as aligned as the most benevolent Buddhist monk – but I (we) feel the community's ontology is too narrow to hold me (us). I'm still not sure what to do about this, so I'm posting the article I wrote for LessWrong – calling for a more 'hippie' rationalism – on Substack instead (actually, there are two articles, and the links are at the end).
Quick intro (who I am)
My name is Melissa. I’m a passionate member of the core AISafety.com team. I was there with the core team when the site was just an empty domain and a loose vision in August 2023. Since then I have co-led, designed, and coded the first official version, and co-led, designed, and coded the second and current version, co-led two of three hackathons, led the creative direction for the current ecosystem map, and I’m currently working to bring lots of more amazing features to our site (stay tuned, it’s going to be awesome)! I’ve also played major roles at AISafety.info, grantmaking.ai, Elicit, and minor roles at Equistamp, PauseAI, and several other well-known EA organizations not directly in the AI safety space.
Okay, I feel a bit cringey writing the above paragraph, but I wanted to build some credibility so I may be more likely to be heard when I point to some problems I see in the AI safety/rationalism/EA space, problems I'm feeling less happy about staying silent about. As someone who generally likes hiding away, this is uncomfortable, but I have decided it’s probably worth the cost.
A volunteer expert in PR at our recent hackathon told me it’s important that I state that my views are my own, and not AISafety.com’s. So, the following views, and any views I express on independent outlets are my own, not AISafety.com’s.
Leaving LessWrong in pursuit of ontological freedom
'Embarkation of the Pilgrims'. Crossing the Atlantic to the new world of Substack.
So, to repeat: there are some problems in the AI safety/rationalism/EA space. If you know the space at all, you know this. I plan on speaking out about these problems in time, granted we have it. This short post will focus on just one meta issue:
It seems like my only chance of reasonably being heard when I speak out about these problems, is if I conform to a narrow and highly specific ontology, but the problems themselves I wish to speak to, in part, come from this ontology.
I want to be clear that I think Yudkowsky is quite brilliant, at least in many ways. I think the sequences are quite brilliant (note I've only read a few), at least in many ways. I think what people are doing on LessWrong has value, and is very reasonable, or at least understandable. I totally empathize with the desire of LessWrongers to want to keep their culture, their ontology. I understand why LessWrong polices comments and posts. I understand LessWrong is not the place to practice a different kind of ontology.
My question then is, where is the place for that?
Okay so here’s a bit more of the backstory. I believe there are some problems in the AI safety/rationalism/EA space (yes, I’m being redundant here on purpose). I wrote an article in March 2024 that I was excited to post on LessWrong. It was meant to be inspiring, to bring people together in a radical, sparkly, kind of way, to shake things up a little, but in a way that would make everyone happier. The tone and ontology were definitely not LessWrong style, although perceivably LessWrong tangential.
I got scared and never posted the article. Scared of, I don’t know. Nit-picky commenters. Being deeply misunderstood. So I take it back, I do know. I was scared of being deeply misunderstood. I shared the article with my mom, and a couple close friends, instead.
Then, this past summer I wrote another article. Similar stuff. I wrote the article with truly pure intentions, a deep desire to help, both individuals and the community. Everything, by anyone’s standards outside of the insularity of the rat community, was quite reasonable, and everything was epistemically humble. But then, I realized I naively fell into the same trap as before. This time though, I really was almost going to post it. Then I got the opinion of two close friends who, like me, linger somewhere on the fringe of rationalism. Both of them explicitly told me, summarized as:
‘If you post this you will get wrecked. You will get not just downvotes but super downvotes. Many. You will likely receive comments so dismissive, that the commenters will not even feel justified in having to explain why they are being dismissive’.
That felt pretty confusing and frustrating to me. Although of course I see some humor in it as well. I do empathize, still. Go up and read the sentence that starts with ‘I totally empathize with the desire of LessWronger’s’, again, if you wish.
I am really, like really, passionate about shaping the next iteration of the universe with as much goodwill, thoughtfulness, [other positive qualities] as possible. I don’t know if anything could possibly matter more to me. And yet, I simply don’t fit as perfectly into the rationalist ontology as rationalists. To try to would be bending myself into a shape I just don’t believe in. One way of thinking of it is: Rationalism and LessWrong are male-pattern autism. I am more like, a mild female-pattern kind of autism. I respect, at a meta-level, the desire to keep LessWrong LessWrong-y. I still would love to see ways, perhaps outside of LessWrong – I’m not sure – to bring more pluralism into the community. To shake things up a bit. For those who are interested, at least interested enough to listen, once. For those who aren’t, it admittedly makes sense to stay on LessWrong only.
I use the word 'we' in the TLDR because I have talked to many other people who peer from the fringes of rationalism and LessWrong – who wish to see the universe unfold to its highest potential in the metaphorical palm of an AI as aligned as the most benevolent Buddhist monk – but who feel the community's ontology is too narrow to hold them. I am very much not alone here. Maybe you are one of these people, in fact (hi!).
So I’ve posted both of my articles on Substack. Where I have rationalist-inspired but not purely rationalist ontological freedom. Where the interface lets me delete mean comments. Here they are:
My second article is titled Call for a more ‘hippie’ rationalism. TLDR: Let’s, if you dare, set aside the comfort we as rationalists find in high-order reasoning, knowing, and what remains in the community of more mainstream cultural norms, and venture into spirituality, and the seemingly wacky, weird, woo-woo territory of the hippie archetype. We’ll inquire, consider, do some thought experiments, see what we can learn, and adjust our positions on the hippie-spiritual-rationalist continuum if we so wish.
Yes, I think rationalism is a vague sociological grouping, which adheres to a general ontology and set of, and also threads of, cultural norms. Mostly inspired by Yudkowsky and The Sequences. I was drawn to EA and then rationalism because 'hippie' alone isn't effective enough, particularly altruistically. In the context of 80 billion animals on factory farms and an impending AI apocalypse I found the hippie space had little to offer. I like that rationalists look for better answers to these problems than 'let's dance around a fire and send our good energy to the universe'. Plus rationalism is good at problem-solving. This is a meta-skill that very much helps a human achieve their goals, including goals like 'I want to build a community in the woods and learn how to dissolve my boundaries and feel one with the forest'. But yeah, if rationalism wasn't so concerned with AI safety, I would take what helped me and stay silent. To say I think AI safety is really important is a massive understatement. Naturally, a healthy AI safety community then, is also important. I think some 'hippie' concepts could bring a lot more health and balance to a community which may possibly decide the fate of the universe (as in, the community may possibly decide the fate of the universe).
TLDR
I (and many others in the community) wish to see the universe unfold to its highest potential in the metaphorical palm of an AI as aligned as the most benevolent Buddhist monk – but I (we) feel the community's ontology is too narrow to hold me (us). I'm still not sure what to do about this, so I'm posting the article I wrote for LessWrong – calling for a more 'hippie' rationalism – on Substack instead (actually, there are two articles, and the links are at the end).
Quick intro (who I am)
My name is Melissa. I’m a passionate member of the core AISafety.com team. I was there with the core team when the site was just an empty domain and a loose vision in August 2023. Since then I have co-led, designed, and coded the first official version, and co-led, designed, and coded the second and current version, co-led two of three hackathons, led the creative direction for the current ecosystem map, and I’m currently working to bring lots of more amazing features to our site (stay tuned, it’s going to be awesome)! I’ve also played major roles at AISafety.info, grantmaking.ai, Elicit, and minor roles at Equistamp, PauseAI, and several other well-known EA organizations not directly in the AI safety space.
Okay, I feel a bit cringey writing the above paragraph, but I wanted to build some credibility so I may be more likely to be heard when I point to some problems I see in the AI safety/rationalism/EA space, problems I'm feeling less happy about staying silent about. As someone who generally likes hiding away, this is uncomfortable, but I have decided it’s probably worth the cost.
A volunteer expert in PR at our recent hackathon told me it’s important that I state that my views are my own, and not AISafety.com’s. So, the following views, and any views I express on independent outlets are my own, not AISafety.com’s.
Leaving LessWrong in pursuit of ontological freedom
'Embarkation of the Pilgrims'. Crossing the Atlantic to the new world of Substack.
So, to repeat: there are some problems in the AI safety/rationalism/EA space. If you know the space at all, you know this. I plan on speaking out about these problems in time, granted we have it. This short post will focus on just one meta issue:
It seems like my only chance of reasonably being heard when I speak out about these problems, is if I conform to a narrow and highly specific ontology, but the problems themselves I wish to speak to, in part, come from this ontology.
I want to be clear that I think Yudkowsky is quite brilliant, at least in many ways. I think the sequences are quite brilliant (note I've only read a few), at least in many ways. I think what people are doing on LessWrong has value, and is very reasonable, or at least understandable. I totally empathize with the desire of LessWrongers to want to keep their culture, their ontology. I understand why LessWrong polices comments and posts. I understand LessWrong is not the place to practice a different kind of ontology.
My question then is, where is the place for that?
Okay so here’s a bit more of the backstory. I believe there are some problems in the AI safety/rationalism/EA space (yes, I’m being redundant here on purpose). I wrote an article in March 2024 that I was excited to post on LessWrong. It was meant to be inspiring, to bring people together in a radical, sparkly, kind of way, to shake things up a little, but in a way that would make everyone happier. The tone and ontology were definitely not LessWrong style, although perceivably LessWrong tangential.
I got scared and never posted the article. Scared of, I don’t know. Nit-picky commenters. Being deeply misunderstood. So I take it back, I do know. I was scared of being deeply misunderstood. I shared the article with my mom, and a couple close friends, instead.
Then, this past summer I wrote another article. Similar stuff. I wrote the article with truly pure intentions, a deep desire to help, both individuals and the community. Everything, by anyone’s standards outside of the insularity of the rat community, was quite reasonable, and everything was epistemically humble. But then, I realized I naively fell into the same trap as before. This time though, I really was almost going to post it. Then I got the opinion of two close friends who, like me, linger somewhere on the fringe of rationalism. Both of them explicitly told me, summarized as:
‘If you post this you will get wrecked. You will get not just downvotes but super downvotes. Many. You will likely receive comments so dismissive, that the commenters will not even feel justified in having to explain why they are being dismissive’.
That felt pretty confusing and frustrating to me. Although of course I see some humor in it as well. I do empathize, still. Go up and read the sentence that starts with ‘I totally empathize with the desire of LessWronger’s’, again, if you wish.
I am really, like really, passionate about shaping the next iteration of the universe with as much goodwill, thoughtfulness, [other positive qualities] as possible. I don’t know if anything could possibly matter more to me. And yet, I simply don’t fit as perfectly into the rationalist ontology as rationalists. To try to would be bending myself into a shape I just don’t believe in. One way of thinking of it is: Rationalism and LessWrong are male-pattern autism. I am more like, a mild female-pattern kind of autism. I respect, at a meta-level, the desire to keep LessWrong LessWrong-y. I still would love to see ways, perhaps outside of LessWrong – I’m not sure – to bring more pluralism into the community. To shake things up a bit. For those who are interested, at least interested enough to listen, once. For those who aren’t, it admittedly makes sense to stay on LessWrong only.
I use the word 'we' in the TLDR because I have talked to many other people who peer from the fringes of rationalism and LessWrong – who wish to see the universe unfold to its highest potential in the metaphorical palm of an AI as aligned as the most benevolent Buddhist monk – but who feel the community's ontology is too narrow to hold them. I am very much not alone here. Maybe you are one of these people, in fact (hi!).
So I’ve posted both of my articles on Substack. Where I have rationalist-inspired but not purely rationalist ontological freedom. Where the interface lets me delete mean comments. Here they are:
Thanks for considering.