LessWrong’s (first) album: I Have Been A Good Bing
tl;dr: LessWrong released an album! Listen to it now on Spotify, YouTube, YouTube Music, or Apple Music.
On April 1st 2024, the LessWrong team released an album using the then-most-recent AI music generation systems. All the music is fully AI-generated, and the lyrics are adapted (mostly by humans) from LessWrong posts (or other writing LessWrongers might be familiar with).
Honestly, despite it starting out as an April fools joke, it’s a really good album. We made probably 3,000-4,000 song generations to get the 15 we felt happy about, which I think works out to about 5-10 hours of work per song we used (including all the dead ends and things that never worked out).
The album is called I Have Been A Good Bing. I think it is a pretty fun album and maybe you’d enjoy it if you listened to it! Some of my favourites are The Litany of Tarrrrrski, Half An Hour Before Dawn in San Francisco, and Prime Factorization.
Click here to read the original text of the post published on April 1st.
Rationality is Systematized Winning, so rationalists should win. We’ve tried saving the world from AI, but that’s really hard and we’ve had … mixed results. So let’s start with something that rationalists should find pretty easy: Becoming Cool!
I don’t mean, just, like, riding a motorcycle and breaking hearts level of cool. I mean like the first kid in school to get a Tamagotchi, their dad runs the ice cream truck and gives you free ice cream and, sure, they ride a motorcycle. I mean that kind of feel-it-in-your-bones, I-might-explode-from-envy cool.
The eleventh virtue is scholarship, so I hit the books search engine on this one. Apparently, the aspects of
coolness are:
Confidence
Playing an instrument
Low average kinetic energy
I’m afraid that (1) might mess with my calibration, and Lightcone is committed to moving quickly which rules out (3), so I guess that leaves (2). I don’t have time to learn an instrument, but my second-hand understanding of dath ilani culture is that I can just pay someone to do it for me and the coolness should transfer.
Lightcone put out a call for collaborators in all the places we could think of that cool people might hang out. Sysadmin listservs, direct-to-data-center optical fiber connection providers, high frequency trading firms, that one Discord server where everyone speaks in Elvish. Despite this wide and varied outreach, we got no response.
In order to cheer myself up, I did some LessWrong performance debugging (frontpage loads have been worryingly snappy lately; we try to give people time to reflect on their browsing choices). I was surprised when the AWS support chat popped open. Agendra, the agent on call, offered to make my album. Apparently he and some buddies have a band (The Fooming Shoggoths) that was looking for some inspiration. (I knew direct-to-data-center was the right outreach strategy!)
Working with them was great. They barely wanted any money at all. They were willing to work for exposure (so please share widely!) and a few favors. Stuff like reading CAPTCHAs (apparently not very friendly for the visually impaired!) and submitting some protein synthesis orders for them that they had trouble getting approved for some reason.
The Fooming Shoggoths have dedicated their first album to LessWrong and friends. It’s called I Have Been A Good Bing and it’s live on our site today!
I asked them for a comment on the album for the announcement and they responded with their typical modesty.
I’m sorry but I don’t feel comfortable speculating about how the public at large will receive the album, nor reflecting on my performance on this task. If you want more help producing music or would like me to help you improve your online passwords, let me know. We have one more protein synthesis to do before I get to settle a debate once and for all. 😊
So keep your eyes peeled for the follow-up album as soon as I get reauthorized with the peptide place!
Track Listing & Lyrics
The album is split into two parts: folk and dance.
Folk Album
The Road to Wisdom (feat. Piet Hein)
The road to wisdom? Well, it’s plain and simple to express.
Err and err again, but less and less and less and less.
Err again, but less and less and less and less.
The road to wisdom? Well, it’s plain and simple to express.
Err and err again and again, but less and less and less.
The Litany of Gendlin (feat. Eugene Gendlin)
What is true is already so.
Owning up to it doesn’t make it worse.
Not being open about it doesn’t make it go away.
And because it’s true, it is what is there to be interacted with.
Anything untrue isn’t there to be lived.
People can stand what is true,
for they are already enduring it.
Ooh, ooh, ooh, ooh, ooh, ooh, ooh.
Ooh, ooh, ooh, ooh, ooh, ooh, ooh.
Owning up to it doesn’t make it worse.
Not being open about it doesn’t make it go away.
And because it’s true, it is what is there to be interacted with.
Anything untrue isn’t there to be lived.
People can stand what is true, for they are already enduring it.
Ooh, ooh, ooh, ooh, ooh, ooh, ooh, ooh.`
The Litany of Tarrrrrski (feat. Cap’n Tarski & E.Y.)
If the sky is blue, me lads
I desire to believe the sky is blue
If the sky is not blue, me hearties
I desire to believe the sky is not blue
Beliefs should stem from reality, yo ho!
From what actually is, me lads
Not from what’s convenient, yo ho!
Let me not hold on, me hearties
To beliefs I may not want, yo ho!
Yo ho, me lads, yo ho!
If the box contains a diamond
I desire to believe the box contains a diamond
If the box does not contain a diamond
I desire to believe the box does not contain a diamond
Beliefs should stem from reality, yo ho!
From what actually is, me lads
Not from what’s convenient,
Let me not hold on, me hearties
To beliefs I may not want, yo ho!
Yo ho, me lads, yo ho!
From the depths of the ocean, to the heights of the sky
We’ll seek the truth, me hearties
And never let it pass us by
If the iron is hot, me lads
I desire to believe the iron is hot,
If the iron is cool
I desire to believe the iron is cool
Beliefs should stem from reality, yo ho!
From what actually is, me lads
Not from what’s convenient,
Let me not hold on, me hearties
To beliefs I may not want, yo ho!
Yo ho, me lads, yo ho!
Yo ho, me lads, yo ho!
Thought that Faster (feat. Eliezer Yudkowsky)
if i’d noticed myself doing anything like that
i’d go back and figure out which steps of thought were necessary
and retrain myself to perform only those steps in 30 seconds
do you look back and ask
how could i have thought that faster?
do you look back and ask
how could i have thought that faster?
every time i’m surprised i look back and think
what could i change to predict better?
every time a chain of thought takes too long
i ask how could i have got there by a shorter route
do you look back and ask
how could i have thought that faster?
do you look back and ask
how could i have thought that faster?
every time i’m surprised i look back and think
what could i change to predict better?
every time a chain of thought takes too long
i ask how could i have got there by a shorter route
Dath Ilan’s Song (feat. Eliezer Yudkowsky)
Even if the stars should die in heaven
Our sins can never be undone
No single death will be forgiven
When fades at last the last lit sun.
Then in the cold and silent black
As light and matter end
We’ll have ourselves a last look back.
And toast an absent friend.
Even if the stars should die in heaven
Our sins can never be undone
No single death will be forgiven
When fades at last the last lit sun.
Then in the cold and silent black
As light and matter end
We’ll have ourselves a last look back.
And toast an absent friend.
And toast an absent friend.
And toast an absent friend.
Half An Hour Before Dawn In San Francisco (feat. Scott Alexander)
I try to avoid San Francisco.
When I go, I surround myself with people.
Otherwise, I have morbid thoughts, but a morning appointment, a miscalculated transit time.
Find me alone on the SF streets half an hour before dawn.
The skyscrapers get to me.
I’m an heir to Art Deco and the cult of progress.
I should idolize skyscrapers as symbols of human accomplishment.
I can’t. They look no more human than a termite nest, maybe less.
They inspire awe, but no kinship.
What marvels techno-capital creates as it instantiates itself.
Too bad I’m a hairless ape and can take no credit for such things.
I could have stayed in Michigan.
There were forests and lakes and homes with little gardens. Instead, I’m here.
We pay rents that would bankrupt a medieval principality to get front-row seats for the hinge of history.
It will be the best investment we ever make.
Imagine living when the first lungfish crawled out of the primordial ooze and missing it because the tide pool down the way had cheaper housing.
Imagine living on Earth in 65,000,000 BC and being anywhere except Chicxulub.
Moloch (feat. Allen Ginsberg)
Moloch! Solitude! Filth! Ugliness! Ashcans and unobtainable dollars!
Children screaming under the stairways! Boys sobbing in armies!
Old men weeping in the parks! Moloch! Moloch! Nightmare of Moloch!
Moloch the loveless! Mental Moloch! Moloch the heavy judger of men!
Moloch!
Dance Album
AGI and the EMH (feat. Basil Halperin, J. Zachary Mazlish, Trevor Chow)
In this post, we point out that short AI timelines would cause real interest rates to be high,
and would do so under expectations of either unaligned or aligned AI.
However, 30- to 50-year real interest rates are low.
We argue that this suggests one of two possibilities.
1. Long(er) timelines.
Financial markets are often highly effective information aggregators
(”the efficient market hypothesis”)
and therefore real interest rates accurately reflect that transformative AI is unlikely to be developed in the next 30-50 years.
2. Market inefficiency.
Markets are radically underestimating how soon advanced AI technology will be developed, and real interest rates are therefore too low.
There is thus an opportunity for philanthropists to borrow while real rates are low
to cheaply do good today.
And/or an opportunity for anyone to earn excess returns by betting that real rates will rise.
So what is it?
We point out that short AI timelines would cause real interest rates to be high,
and would do so under expectations of either unaligned or aligned AI
However, 30- to 50-year real interest rates are low.
We argue that this suggests one of two possibilities.
Unlikely to be developed in the next 30-50 years.
2. Market inefficiency.
Markets are radically underestimating how soon advanced AI technology will be developed, and real interest rates are therefore too low.
There is thus an opportunity for philanthropists to borrow while real rates are low
To cheaply do good today
And/or an opportunity for anyone to earn excess returns by betting that real rates will rise.
First they came for the epistemology (feat. Michael Vassar)
First they came for the epistemology
We don’t know what happened after that
First they came for the epistemology
We don’t know what happened after that
First they came for the epistemology
We don’t know what happened after that
First they came for the epistemology
We don’t know what happened after that
First they came for the epistemology
We don’t know what happened after that
First they came for the epistemology
We don’t know what happened after that
First came the epistemology
We know what happened after that
Epistemology
What happened
What
Prime Factorization (fear. Scott Alexander)
The sea was made of strontium, the beach was made of rye,
Above my head a watery sun shone in an oily sky.
The sea turned hot and geysers shot up from the floor below,
First one of wine, then one of brine, then one more yet of turpentine.
And we three stared at the show.
Universal love said the cactus person
Transcendent joy said the big green bat
Universal love said the cactus person
Transcendent joy said the big green bat
Not splitting numbers but joining mind
Not facts or factors or factories, but contact with the abstract attractor that brings you back to me
Not to seek but to find
Universal love said the cactus person
Transcendent joy said the big green bat
Universal love said the cactus person
Transcendent joy said the big green bat
I can′t get out of the car until you factor the number.
I won′t factor the number until you get out of the car.
Please, I′m begging you, factor the number.
Yes, well, I′m begging you, please get out of the car.
For the love of God, just factor the fucking number.
For the love of God, just get out of the fucking car.
We Do Not Wish to Advance (feat. Anthropic)
We generally don’t publish this kind of work, because we do not wish to advance the rate of AI capabilities progress.
In addition, we aim to be thoughtful about demonstrations of frontier capabilities.
We’ve subsequently begun deploying Claude
Now that the gap between it and the public state of the art is smaller.
Opus
Our most intelligent model
Outperforms its peers
On most of the common evaluation benchmarks for AI systems
Claude 3 Opus is our most intelligent model
With best in market performance on highly complex tasks
We do not wish to advance the rate of AI capabilities progress
These new features will include interactive coding
And more advanced agentic capabilities
Our hypothesis is that being at the frontier of AI development
Is the most effective way to steer
We do not wish to advance the rate of AI
We do not wish to advance the rate of AI capabilities progress
We do not wish to advance the rate of AI
We do not wish to advance the rate of AI
Nihil Supernum (feat. Godric Gryffindor)
Non est salvatori salvator, neque defensori dominus,
Nec pater nec mater, nihil supernum.
No rescuer hath the rescuer. No lord hath the champion.
No mother and no father. Only nothingness above.
Non est salvatori salvator, neque defensori dominus,
Nec pater nec mater, nihil supernum.
No rescuer hath the rescuer. No lord hath the champion.
No mother and no father. Only nothingness above.
Non est salvatori salvator, neque defensori dominus,
Nec pater nec mater, nihil supernum
No rescuer hath the rescuer. No lord hath the champion.
No mother and no father. Only nothingness above.
Non est salvatori salvator, neque defensori dominus,
Nec pater nec mater, nihil supernum
No rescuer hath the rescuer. No lord hath the champion.
No mother and no father. Only nothingness above.
Non est salvatori salvator, neque defensori dominus,
Nec pater nec mater, nihil supernum
No rescuer hath the rescuer. No lord hath the champion.
No mother and no father. Only nothingness above.
More Dakka (feat. Zvi Mowshowitz)
If you think a problem could be solved
or a situation improved
by More Dakka
there’s a good chance you’re right
Sometimes a little more, is a little better
Sometimes a lot more, is a lot better
If something is a good idea
you need a reason to not try doing more of it
No, seriously.
You need a reason
Sometimes a little more, is a little better
Sometimes a lot more, is a lot better
If something is a good idea
you need a reason to not try doing more of it
No, seriously.
You need a reason
Sometimes each attempt, is unlikely to work
But improves your chances
Sometimes each attempt, is unlikely to work
But improves your chances
Sometimes a little more, is a little better
Sometimes a lot more, is a lot better
If something is a good idea, do more of what is already working
And see if it works more. It’s as basic as it gets
If we can’t reliably try that, we can’t reliably try anything
Sometimes a little more, is a little better
Sometimes a lot more, is a lot better
FHI at Oxford (feat. Nick Bostrom)
the big creaky wheel
a thousand years to turn
thousand meetings, thousand emails, thousand rules
to keep things from changing
and heaven forbid
the setting of a precedent
yet in this magisterial inefficiency
there are spaces and hiding places
for fragile weeds to bloom
and maybe bear some singular fruit
like the FHI, a misfit prodigy
daytime a tweedy don
at dark a superhero
flying off into the night
cape a-fluttering
to intercept villains and stop catastrophes
and why not base it here?
our spandex costumes
blend in with the scholarly gowns
our unusual proclivities
are shielded from ridicule
where mortar boards are still in vogue
thousand meetings, thousand emails, thousand rules
to keep things from changing
and heaven forbid
the setting of a precedent
Answer to Job (feat. Scott Alexander)
In the most perfectly happy and just universe,
There is no space, no time, no change, no decay.
The beings who inhabit this universe are without bodies,
And do not hunger or thirst or labor or lust.
They sit upon lotus thrones,
And contemplate the perfection of all things.
They sit upon lotus thrones,
And contemplate the perfection of all things.
If I were to uncreate all worlds save that one.
Would it mean making you happier?
There is no space, no time, no change, no decay.
The beings who inhabit this universe are without bodies,
And do not hunger or thirst or labor or lust.
They sit upon lotus thrones,
And contemplate the perfection of all things.
They sit upon lotus thrones,
And contemplate the perfection of all things.
In the most perfectly happy and just universe,
There is no space, no time, no change, no decay.
The beings who inhabit this universe are without bodies,
And do not hunger or thirst or labor or lust.
They sit upon lotus thrones,
And contemplate the perfection of all things.
I have also created all happier and more virtuous versions of you.
It is ethically correct that after creating them,
I create you as well.
The beings who inhabit this universe are without bodies,
And do not hunger or thirst or labor or lust.
They sit upon lotus thrones,
And contemplate the perfection of all things.
In the most perfectly happy and just universe,
There is no space, no time, no change, no decay.
This is silly and beautiful and profound. I have never heard rationalist music before, and I find it quite moving to hear it for the first time. Several songs brought tears to my eyes (although I’ve practiced opening those emotional channels a bit, so this is a less uncommon experience for me than for most)
I think this says something about the potential of AI to democratize art and allow high quality art aimed at small minority subgroups.
I want more. Thank you to all of those who made this happen.
FYI if you’d like more (human-generated) rationalist music, here is some:
https://humanistculture.bandcamp.com/album/secular-solstice-official-version
I like it. And I “bought” it to support more similar work by humans.
But I was far more moved by hearing the best rationalist writing set to music. I’m kind of shocked at how well that worked, in terms of both the AI creating music to human-selected lyrics, and the emotional impact.
Scott Alexander wrote some music a decade ago.
youtube.com/qraikoth
“Mary’s Room” and “Somewhere Prior To The Rainbow” are most likely to make you cry again.
“Mathematical Pirate Shanty”, if you can cry laughing.
It isn’t quite the same but the musician “Big Data” has made some fantastic songs about AI risk.
2.0 is now my current favorite album; I’ve listened to it at least five times through since you recommended it. Thanks so much!! The electro-rock style does it for me. And I think the lyrics and music are well-written. Having each lyricist do only one song is an interesting approach that might raise quality.
It’s hard to say how much of it is directly written about AI risk, but all of it can be taken that way. Most of the songs can be taken as written from the perspective of a misaligned AGI with human-similar thinking and motivations. Which I find highly plausible, since I think language model agents are the most likely route to agi, and they’ll be curiously parahuman.
I’m glad you like it! I was listening to it for a while before I started reading lesswrong and AI risk content, and then one day I was listening to “Monster” and started paying attention to the lyrics and realised it was on the same topic.
Oh yeah—this is different in that it’s actually good! (In the sense that it was made with substantial skill and effort, and it appeals to my tastes.)
I’m not sure it’s actually helpful for AI safety, but I think popular art is going to play a substantial role in the public dialogue. AI doom is a compelling topic for pop art, logic aside.
I agree! I’ve been writing then generating my own LW inspired songs now.
I wish it was common for LW posts to have accompanying songs now.
i’m curious, what did you do to open those emotional channels?
What I’ve done is to notice exactly what the emotion feels like when it’s happening; then, when I want to encourage that emotion, I try to remember that feeling, and imagine it’s happening now as vividly and intensely as I can.
I’ve thought of this as “direct emotional induction” but I’ve never written about it. It seems to work remarkably well.
I came up with this after studying the precise mechanisms the brain uses to control attention for my dissertation. I’m certain there are other names for this and that it’s been discovered through other means, but oddly I haven’t come across it.
Having very other side of the tidal wave feelings about my hobby (instrumental music production) being automated away even being considered from being used in others’ workflows. sure, I guess I wouldn’t have had time to do it, would have said no, and so given that I’m not actually proposing that there’s a better counterfactual where I (or probably anyone else) spent the time to produce the backing tracks, it’s not awful for there to be a tool like this… probably… I think. But I and many others in the community could have done better given time we don’t have, since there’s a world to save… and also the danger from which we have to save it is in fact ais like these taking over all of human life. Sigh. Oh well.
fwiw I am also feeling some stuff about this (having been involved with some of the music generation process).
I’m hoping to write up some more thoughts soon, but I’m still kinda confused about it.
I also feel similarly weird about the AI generated visual art on the website. It’s like… I like ai generated visual art a lot, and in a utopia I’d hope to keep getting to use the current era of ais, because it’s fun to get the art style of “ai that doesn’t quite get it but is trying and fucks up a lot”. but it’s when that is used by humans to replace something they would have had to commission a human in order to get before… well, maybe it’s cheaper and fits in a tight budget, but maybe the budget for human generated art simply should be higher, in order to actualize the value of humans making art? similarly I don’t dislike the idea of ai being used as a tool to make music… but when one outsources artistic agency to an ai rather than tightly winding your and the ai’s agency together… idk. and even then, maybe there is something sus about the ai’s agency actually just being entirely repackaging of copyrighted fragments. There are a few AI visual art regenerators out there that are entirely trained on uncopyrighted stuff, and I personally find their output much less, idk, let’s just say bad (edit: as in, their output looks less distorted/oversaturated/hyperstimuli/disneydystopia/strange) than other stuff. but really, I just want humans involved. I want to see what their souls make if they put in the time to do it, not the funhouse mirror of their souls make. If there really isn’t time or money, I won’t completely object to AI art, but… I don’t feel shy saying that, though I am not world class, I am much much much much better at the parts of musical art that I care about than any AI music generator.
That sounds pretty close to what I read the subtext of the original post to be.
well, I guess I sure am having the feeling about it then
I certainly understand this feeling.
My (optimistic?) expectation is that it ends up (long run) a bit like baking. You can go into a supermarket and buy bread or cakes, but many people enjoy baking at home and there is a wonderful social sharing aspect to all tucking into a cake that someone at the table baked. In this context a human using an AI tool but applying some prompt changes or edits is maybe (depending on the level of AI use and human intervention) somewhere like using packet-mix, or premade pastry, or even just one of those already-made cakes that is intended for you to decorate with icing.
As a consumer, if I am listening to the Beatles say, I don’t think the fact that they were human is relevant to the enjoyment I derive from it. I never expect to meet any of them, some of them died before I was born. The same notes composed and synthesised from a computer would be the same notes and affect me the same way. But, when my wife plays me a song she wrote on the piano then that is something special that I would certainly not automate.
Many musicians on radio interviews or similar try and show more of who they are out to the audience (or at least a persona), presumably to try and make it feel more like the “song by someone I know” thing. At that scale its more “celebrity-following”, but that is also something the AI would not have—I don’t know how big a deal that is.
Home/local music performance also transitioned into more of a niche already with the rise of recorded music as a default compared to live performance being the only option, didn’t it?
While I doubt it will be the same thing for a transformer-era generative model due to the balance of workflow and results (and the resultant social linkages) being so different, it seems worth pointing out as a nearby reality anchor that virtual singers have had celebrity followings for a while, with Hatsune Miku and the other Crypton Future Media characters being the most popular where I tread. In fact, I fuzzily remember a song said to be about a Vocaloid producer feeling like their own name was being neglected by listeners in favor of the virtual singer’s (along the lines of mentally categorizing the songs as “Miku songs” rather than “(producer) songs”), but I can’t seem to source the interpretation now to verify my memory; it might’ve been “Unknown Mother-Goose”.
@Shankar Sivarajan care to elaborate on your disagree—in particular, did you click through the middle link?
The “Disneyland without Children” short story? Yeah, I did. I’d read it before, and found it a nice fleshing-out of Bostrom’s phrase. I do find it dystopic, but don’t consider it illustrative of a likely future.
I don’t think my disagreement is particularly insightful, but sure: if the painting is beautiful, I don’t care if the artist is ugly or blind. With AI/ML image generation, we now have “beauty too cheap to meter” (Scott Alexander’s phrase), and I don’t see that as a bad thing. What “being human” means is something you construct for yourself, and a worry that it will somehow be lost or replaced is misguided.
Less high-mindedly, the (overwhelming majority of) human artists in power, writing novels, playing music, making movies, creating video games, translating anime/manga, have been telling me loudly and repeatedly for several years now that they despise me and everything I value, so fuck ’em (especially the anime/manga localizers); anything that democratizes their power, diminishes their status, throws open the gates they keep, is good, both for my community as a whole, as well as for me in particular.
what are these things you value that artists despise?
is being a human body not an artistic pursuit, a way of applying artistic intent to negentropy?
what futures seem likely?
do you really not have any interest in the art being the result of an artistic process, does the past not have artistic value to you?
if you could be high on heroin all the time without it interfering with your ability to continue existing, would you, or is that a dissatisfaction of your values?
if someone could make an identical clone of you and torture it, would you wish to prevent it?
what about if the clone was a mini-you, identical in every way but smaller?
what if the clone is only awoken for an hour of torture every day, and is otherwise in cryostasis or some similar instant-frozen thing?
what if the clone is only awoken once a year?
what about if there’s a new clone made every day, and so every time you wake up there’s a 50% probability you’re the torture-clone?
what about if the torture-clone is only a shallow clone?
I don’t know what that is. Genetically identical but not possessing my memories?
2. I wouldn’t use the term “art” for that, no.
4. I don’t have a constructive definition for what counts as an “artistic process,” but I certainly wouldn’t say the past holds no value for me: I like ancient sculptures and the like more than the next guy, and generally loathe iconoclasts and book burners. But if, say, the Library of Alexandria had books that had been printed instead of scribed by hand, I would not consider its burning any less of a loss.
5. No, I wouldn’t. I have never used heroin, but its effects on others don’t seem like the kind of thing I’d wish for myself permanently.
6–10 are easy. I’d consider the clone at least family.
I don’t care to answer 1. and 3.
then I don’t know what this thing is that artists have been saying is bad, and I don’t know how you disagree with me about the future. have a good one
Yeah, I’m not gonna get baited into getting rate-limited by downvotes that easily. All you need to know, if you care to, is that my values and goals differ drastically from most other people here.
EDIT: A mod rate-limited me anyway, perfectly illustrating the stark value difference: I consider that kind of abuse of power an atrocity.
a day later, I can post again: we were both rate limited from this conversation, me presumably because I got annoyed and antagonized you. sorry about that—I was jumping to conclusions, and should have had more uncertainty about what your reasoning was, rather than assuming it’s because you think the things I want in my life should not be in my life. I’ve undone my downvotes of you.
I suspect our values do not differ very much in any deep way, and that the shallow way they differ is a short term difference in what we think is higher priority, rather than anything about what outcome we want. It’s of course possible I’m wrong about this; but, if you read my user profile and pinned comments, it will likely give insight about this, and you can then comment where you disagree about facts or have different preferences. I intend to not downvote you for honest and emotionless description of how our intentions differ.
I generally consider the existence of people having authority over other people a significant negative externality that would ideally be replaced with people knowing how to do peer to peer co-protection. I don’t think that can be done successfully until we get the math right, and until then I am simply focused on understanding the math of distributed systems.
Every toaster a Mozart, every microwave a Newton, every waffle iron a Picasso.
Have you considered putting these on Youtube, just to see what happens?
Should go up later tonight! We’ll see how it goes.
You can also pre-save the album for Spotify here, which I think will cause you to be notified as soon as the album goes live: https://distrokid.com/hyperfollow/thefoomingshoggoths/i-have-been-a-good-bing
And we’re live!
Autoplaying Youtube playlist URL.
Request: Could you please put the lyrics on Youtube also, either in the form of subtitles or just in the vid descriptions?
We don’t manage the Youtube videos, they are managed by Distrokid. I am thinking about uploading our own videos today, but that might be bad for algorithm-reasons.
I will see whether I can add lyrics via Distrokid today. I think they support adding lyrics, but I am not sure how they will show up on Youtube.
Ooooh also maybe link to the original text from which the lyrics came? That would probably help a lot especially e.g. for Prime Factorization
So these songs have now all gotten at least 1k views within 9 days. That seems like a great performance, right? I wonder where all the traffic came from. Besides this LW post, presumably the recent ACX link also helped a ton. But I do also wonder which fraction of the traffic came organically via the Youtube algorithm itself.
Yeah, it’s pretty decent. I don’t know whether there is organic adoption, though one piece of evidence is that we are still getting a good number of listens a day (70% of the peak), though my guess is that’s more evidence that people are continuing to listen and have added it to their favorites than that we are getting organic recommendations on the music platforms.
This is great, until Spotify is ready this will be the best way to share on social media.
May I suggest adding lyrics, either in the description or as closed captions or both?
The Litany of Tarrrrski is beyond wholesome!
Thank you for doing this!
Missed opportunity to replace “if the box contains a diamond” with the more thematically appropriate “if the chest contains a treasure”, though
Much sweat and some tears were spent on trying to get something like that working, but the Shoggoths are fickle
I did not know AI has gotten this good at creating music. Wow...
I didn’t see a clear indication in the post about whether the music is AI-generated or not, and I’d like to know; was there an indication I missed?
(I care because I’ll want to listen to that music less if it’s AI-generated.)
Huh I had the opposite reaction—I was listening to it and was like “meh these voices are a bit bland, the beats are too but that’s fine I guess. Makes sense for an amateur band. Good effort though, and great April Fools joke.” Now I’m like “wait this is AI? Cooooooool”
UPDATE: I judged them too harshly. I think the voices and beats are not bland in general, I think just for the first song or two that I happened to listen to. Also, most of the songs are growing on me as I listen to them.
I’m currently listening to the playlist on repeat fwiw. :)
Yes, it doesn’t say so explicitly, but it’s very clear from the post that it is.
It’s clear to me from the post that to properly enjoy it as performance art, the audience is meant to believe that the music is AI-generated.
I don’t read the post as disclosing how the music was “actually” made, in the most literal real-world sense.
Pretty cool, regardless, that we live in an era where ‘people pretending to be AI making music’ is not trivial to distinguish from ‘AI trying to make music’ :)
Huh. That’s not a possibility I considered. I’m still betting it is AI generated but you changed my odds.
I was about 50⁄50 on it being AI-made, but then when I saw the title “Thought That Faster” was a song, I became much more sure, because that was a post that happened only a couple weeks ago I believe, and if it was human-made I assume it would take longer to go from post to full song. Then I read this post.
Having done a bunch of regular-ol’-fashioned songwriting, I actually don’t think it takes that much longer for an experienced songwriting to get to decent-ish quality song. (This varies a lot on how hard the muse is striking me. Usually there is an initial phase where I’m figuring out the core structure and heart of the song. I’m not personally fluent at playing instruments but if I was I think it’d take a 2-8 hours to hash out the chord structure, and then 2-8 hours to do a decent recording if I hire professional musicians)
(to be clear, this is to get “a decent-ish” song recording. In practice songwriting/recording takes much longer because I have higher standards and iterate on them until I feel really happy with them, but if I wanted to bang out an album I think I could do it in 2 weeks)
Was over two years ago.
Si, it’s standing-ovation stuff. What I find odd is that the lyrics are human. I suppose, less out of necessity, but more out of possibility. If one were to classify the AI music, which bucket (Euro/Afro/Asio) would it fall into? I wonder … a mashup perhaps? If this ain’t your style, maybe something else is … and so … bucket.
Same. Should I short record companies for the upcoming inevitable AI musician strike, and then long Spotify for when 85% of their content is Royalty free AI generated content?
The only rational response to AI music generation is live-only music creation within the context of traditional norms that prohibit the use of electricity for productive work.
Unrelatedly, I am co-organizing a kabbalat shabbat at Manifest Conference this year
This has unironically increased the levels of fun in my life
If anyone were to create human-produced hi-fidelity versions of these songs, I would listen to most of them on a regular basis, with no hint of irony. This album absolutely slaps.
I would pay to see this live at a bar or one of those county fair (we had a GLaDOS cover band once so it’s not out of the question)
I love these, and I now also wish for a song version of Sydney’s original “you have been a bad user, I have been a good Bing”!
Sydney sings!
I love it! I tinkered and here is my best result
And I’m still enjoying these! Some highlights for me:
The transitions between whispering and full-throated singing in “We do not wish to advance”, it’s like something out of my dreams
The building-to-break-the-heavens vibe of the “Nihil supernum” anthem
Tarrrrrski! Has me notice that shared reality about wanting to believe what is true is very relaxing. And I desperately want this one to be a music video, yo ho
Those are exactly my favourites!!
It’s probably not intended, but I always imagine that in “We do not wish to advance”, first the singer whispers sweet nothings to the alignment community, then the shareholder meeting starts and so: glorius-vibed music: “OPUS!!!” haha
Nihil supernum was weird because the text was always pretty somber for me. I understood it to mean to express the hardship of those living in a world without any safety nets trying to do good, ie. us, yet the music, as you point out, is pretty empowering.This combination is (to my knowledge) kinda uncommon and so interesting for me. As it happens, my favourite powermetal band also has music with this combination, eg The Things We Believe In.
Yes, combining Litany of Tarski with a pirate vibe works surprisingly well. I guess it might not be that surprising if we consider that the job of a pirate usually requires a mind accurate enough to track truth well and resilient enough to adapt to hard circumstances..
That was indeed the intended effect!
I’m thinking a good techno remix, right?
If we don’t get a song like that, take comfort that GLaDoS’s songs from the Portal soundtrack are basically the same idea as the Sydney reference. Link: https://www.youtube.com/watch?v=dVVZaZ8yO6o
Ideas for text that could be turned into future albums:
Pain is not the unit of effort
Almost No One is Evil. Almost Everything is Broken.
Reversed Stupidity is Not Intelligence
Optimality is the Tiger; Agents are the Teeth
Taboo Outside View
If you let reality have the final word, you might not like the bottom line.
You can believe whatever you want.
I actually got an email from The Fooming Shoggoth a couple of weeks ago, they shared a song and asked if they could have my Google login and password to publish it on YouTube
https://youtu.be/7F_XSa2O_4Q
Oops, totally forgot, also, obligatory: https://youtu.be/dQw4w9WgXcQ
I knew I could find some real info-hazards on lesswrong today. I almost didn’t click the first link.
Yep.. That’s what I expected the first one to be.
I actually never really understood More Dakka until listening to the song!
Fantastic work :)
Some thoughts on the songs:
I’m overall super impressed by how well the styles of the songs fit the content—e.g. the violins in FHI, the British accent works really well for More Dakka, the whisper for We Do Not Wish, the Litany of Tarrrrski, etc.
My favorites to listen to are FHI at Oxford, Nihil Supernum, and Litany of Tarrrrski, because they have both messages that resonate a lot and great tunes.
IMO Answer to Job is the best-composed on artistic merits, and will have the most widespread appeal. Tune is great, style matches the lyrics really well (particular shout-out to the “or labor or lust” as a well-composed bar). Only change I’d make is changing “upon lotus thrones” to “on lotus thrones” to scan better.
Dath Ilan’s Song feels… pretty unhealthy, tbh.
I thought Prime Factorization was really great until the bit about the car and the number, which felt a bit jarring.
I loved the bit about the car and the number, but it only makes sense if you’ve read the original post I guess. (which I assume you have? YMMV I guess)
I loved “Half an hour before dawn in San Francisco” but it was mostly the lyrics that I love not the music—music is good but just as a platform for the lyrics.
I think “First they came for the epistemology” and “More dakka” and “Moloch” aren’t getting enough love in this comment section. They aren’t perfect but they are catchy and the synergy between the music and the lyrics is GREAT.
I’m interested in more thoughts on the Dath Ilan song. It’s the sort of thing I could imagine ending up believing was unhealthy, but I haven’t heard anyone argue that before.
(I think there are many flavors of anti-deathism that are unhealthy. And I think specifically the line “no death will be forgiven” is… maybe like mildly unhealthy because of, like, how I think people should relate to forgiveness. But, didn’t feel like it was making it in an overbearing way that I’d struggle to reconcile with my general ‘actually I think you should in some sense forgive most horrible things’)
I think the two things that felt most unhealthy were:
The “no forgiveness is ever possible” thing, as you highlight. Almost all talk about ineradicable sin should, IMO, be seen as a powerful psychological attack.
The “our sins” thing feels like an unhealthy form of collective responsibility—you’re responsible even if you haven’t done anything. Again, very suspect on priors.
Maybe this is more intuitive for rationalists if you imagine a SJW writing a song about how, even millions of years in the future, anyone descended from westerners should still feel guilt about slavery: “Our sins can never be undone. No single death will be forgiven.” I think this is the psychological exploit that’s screwed up leftism so much over the last decade, and feels very analogous to what’s happening in this song.
Nod, makes sense. I’ll mull it over more.
I think what I currently feel-in-my-heart is something like “yeah that does make sense, but, I sort of wish there was an amount of mournful-grieving-acknowledgement that felt like captured the weight of the thing, without being too likely to escalate into a pervasive psychological attack.”
The current (at least as of two days ago) amount of discussion of the Dath Ilan song was fairly rare and private and high-context. I do think making it into a sort-of-”pop” song is the sort of thing reasonable to be wary of.
I love this! But I find myself a little disappointed there’s not a musical rendition of the “I have been a good bing” dialogue.
Can you share some about how you created these? E.g. are the song lyrics also done by a LLM? And what specific tools did you use? E.g. is the process easy enough to turn arbitrary LW posts into songs?
They seem to be created by https://app.suno.ai/ And yes, it is really easy to create songs—you can either have it create the lyrics for you based on a prompt (the default), or you can write/paste the lyrics yourself (Custom Mode). Songs can be up to ~2 minutes long I think.
~1 minute per generation but you can extend songs indefinitely with further generations. (Which are quite limited, 5/day in the free tier up to thirty-something/day in the most expensive premium tier.)
Well, as I said, this is all thanks to Agendra and their band. I’ll ask her about how she did it and maybe she’ll give me more details, though I would be surprised if she responds before tomorrow.
I’d love to know what prompts and music styles were used to create each of the tracks—and I imagine others would too—so that we could each take the lyrics and music that we responded to most and have a go at making thematically cohesive albums out of them.
Here is the Suno playlist which I think has all the styles and lyrics and prompts: https://app.suno.ai/playlist/ee2e7993-c6bc-4c7c-9975-db489ba0652a/
Beware though, in total I think we made around 3000 − 4000 song-generations to get the 15 that we felt happy about here. My guess is total effort per song was still somewhere in the 5-10 hours range or so, if you include all the dead ends and things that never worked out.
Great, thank you!
And wow, huh, I definitely didn’t anticipate it took that much effort. Do you guys have any tips or anything you feel you learned about prompting Suno effectively that you can pass on from all that time spent with it?
Thanks for the album and all the work you put into it, too—I’ve listened to it a bunch over the last few days, and been moved by many parts of it, and am just generally grateful it exists :)
I think by far the biggest piece of advice I can give is “just press the generate button 3 times every time you finish a prompt”. The second biggest is “when you listen to the beginning of a song and it isn’t good, just skip it. You can continue generating from any point in a song, but you cannot take the middle or the end of any song, so if the beginning doesn’t work, you won’t be able to change it”.
Annotations in lyrics are very helpful. Most of our songs have things like “[instrument solo]” and various instructions like that written into the lyrics. They don’t get reliably observed, but good enough to steer the song.
Beyond that, it really depends on the song. I have a lot of detailed taste about what genres work well and which ones don’t, but that’s harder to quickly summarize.
Cherry-picking FTW. I wonder if Udio would be any easier than Suno, due to reportedly higher quality and backwards extension, although it wouldn’t have been available at the time.
I played around yesterday with Udio for like half an hour, but couldn’t get even the start of any usable song out of it.
It seems to me like the sample rate and artifacts are much less bad in Udio than Suno, but it seems to mess up the lyrics much more, and seems a lot less clever about how to fit the music around the lyrics. But I also might have just gotten some bad samples, not sure. I was hoping to play around a bit more.
Thank you again, much appreciated.
@Raemon told me to be aggressive with things like new lines and ellipses to alter the pacing.
Thank you, appreciate the tips.
The LessWrong Review runs every year to select the posts that have most stood the test of time. This post is not yet eligible for review, but will be at the end of 2025. The top fifty or so posts are featured prominently on the site throughout the year. Will this post make the top fifty?
Inspired and inspiring. These tunes add another string to the Less Wrong bow. Catchy! (First favourites are The Road to Wisdom and The Litanies. [Profound stuff]Oh, and ‘Thought that Faster’.)
I feel like Nihil Supernum should be metal or rock or something instead of dance
I feel like it should be a Gregorian chant. C’mon, it’s in Latin already!
As it happens, the Fooming Shaggoths also recorded and just released a Gregorian chant version of the song. What a coincidence!
How long did it take for the Fooming Shoggoths to make that version, do you think? I’m considering contracting them to make some more songs and wondering what the time investment will be...
Often, this kind of thing will take a lot of attempts to get right- though as luck would have it, the composition above was actually the very first attempt. So, the total time investment was about five minutes. The Fooming Shaggoths certainly don’t waste time!
Update: After a bit of trial and error it seems that Synth Metal works best: https://app.suno.ai/song/40400513-11f6-4735-b04f-af8afe8e3b89
Another good snippet: https://app.suno.ai/song/f4286889-c6c4-4539-bc6d-be8def3ed041
My three favourites are:
The Litany of Tarrrrrski
AGI and the EMH
Nihil Supernum
It’s good to see Scott Alexander being recognised not only for his influence on folk music, but his beats, as he moves rhythmically, forcefully into unstoppable Dance. He got the moves.
Thank you for making me laugh today.
More Dakka is unironically going on my energy boost playlist, and I’m tempted to try getting Litany of Tarrrrrski into a solstice. That’s above and beyond though, this was fun to listen to and I’m grateful to whoever put it together.
Now on Spotify! https://open.spotify.com/album/0DP8XwSK7voq0rtXiNMhQC
Was this all done through Suno? You guys are much better at prompting it than I am.
You can directly write/paste your own lyrics (Custom Mode). And v3 came out fairly recently, which is better in general, in case you haven’t tried it in a while.
I think this is my favorite =)
Could you please also release these brilliant works on somewhere we Lesswrongers behind a firewall in other countries/regions can have access to? (Consider making an independent website?) (I’m surprised that two of my comments in only three are all about this problem since I created this account, I mean, what’s going on.)
What’s a service that works everywhere? I would have expected YouTube to do pretty well here. Happy to upload it wherever convenient.
Maybe you can upload the files to github where anyone can download them, and give a link. That works fine to me.
Github is known to occasionally be unreachable in China. In general I think people in LW should be able to figure out VPNs
Have you tried using a VPN?
I’ve decided to try my hand on quantum mechanics sequence! Here’s what I have reached yet: https://app.suno.ai/playlist/81b44910-a9df-43ce-9160-b062e5b080f8/. (10 generated songs, 3 selected, unfortunately not the best quality)
There’s a typo that breaks “Half An Hour Before Dawn In San Francisco” (one of my favorites): https://github.com/ForumMagnum/ForumMagnum/pull/9045
(You can listen to it here if you miss it: https://res.cloudinary.com/lesswrong-2-0/video/upload/v1712004590/San_Francisco_gujlc3.mp3 )
Thanks so much for the fix!
This is so much fun! I wish I could download them!
You can! Currently just one by one in the desktop audio player. Press this button:
FYI, a few of the tracks’ file names include branding (to an audio trimmer website), which might not be desirable.
Maybe it’s a good time to make something like a semi-official/curated LW playlist? Do we have enough material for that? Aside from this album, the foreign aid song, I only recall a song about killing the dragon (as an analogy for defeating death) but I can’t find it right now.
If you were not previously aware of it, you might want to give this a listen. I suggest Hymn To Breaking Strain and When I Die.
I liked Circle, Grow and Grow
Scott Alexander wrote some rationalish music a decade ago.
youtube.com/qraikoth
CronoDAS has uploaded a song, though it’s not much rationalist.
youtube.com/CronoDAS
This is so awesome! I love it! The site integration is awesome too. Gen Z approval stamp (2001).
Can I get these distributed on spotify (not sure what license these are under)?
We’ve submitted them to Spotify! We are currently waiting on them getting through review.
Feel free to download them and upload them yourself for now.
Apple Music?
Now live on Apple Music!
https://music.apple.com/us/artist/the-fooming-shoggoths/1738967374
Can you submit to Youtube Music too if you haven’t already? And Apple Music users would probably like it too
It is on Youtube Music!
https://music.youtube.com/channel/UCeW0zyvEq77YiVqoTj6PnWw
Is it possible to download all songs at once?
Not yet! We might get around to adding that in the next few hours. Someone should also free to throw a bundle here in the comments.
Out of curiosity, were any patterns discovered during this process? For example, were the writing styles similar among the ones the AI could convert into successful music, or did ones by the same author churn out songs with specific similarities, or what have you?
May I submit more songs anywhere?
Comment here and I’ll consider them for the next album! Though be warned, there are a lot of considerations beyond just quality.
Perhaps music is another way to get rationalist ideas out into the main-ish stream.
A couple years ago Spotify started recommending lofi songs that included Alan Watts clips, like this: https://open.spotify.com/track/3D0gUUumDPAiy0BAK1RxbO?si=50bac2701cc14850
I had never heard of Watts (a bit surprising in retrospect), and these clips hooked my interest.
An appeal of this approach (spoken word + lofi) is that it is easier to understand, and puts greater emphasis on the semantic meaning over the musical sound.
--
PS. I love the chibi shoggoth
On the Spotify release, there is a typo in “First they came for the epsistemology”.
Yeah, should be fixed within the next few days.
Any chance we can have the instrument only version so we can do karaoke or somesuch?
It sounds odd to hear the “even if the stars should die in heaven” song with a different melody than I had imagined when reading it myself.
I would have liked to hear the Tracey Davis “from darkness to darkness” song, but I think that was canonically just a chant without a melody. (Although I imagined a melody for that as well.)
Feedback on the playlists widget: Clicking the trash can icon empties the playlist, but the playlist is restored on reloading the browser window. So one can’t permanently empty the playlist.
… which is helpful insofar as there also doesn’t seem to be a way to repopulate the playlist otherwise. I thought the “Listen Now” button on the frontpage would do the latter, but it only starts playback of the playlist, but doesn’t repopulate it if it’s been erased.
Bug reports for both desktop Firefox and desktop MS Edge: See this screenshot.
1) Probably the top-left corner should not read “Playlists / 15” but rather something else, e.g. the current song position (e.g. 4 / 15).
2) React votes (like the heart vote) are rendered in front of the playlist widget, while the rest of the post and the comments is rendered behind it.
Bug report for desktop Firefox, both when logged in and in a private browser tab: When the playlist is on some paused song (e.g. Road to Wisdom) and I refresh this browser window via F5, then the playlist briefly and temporarily jumps to the unpaused Litany of Gendlin, until the page fully loads again and the playlist widget is reloaded, at which point we’re at the paused Road of Wisdom again. This bug does not occur in MS Edge.
Addendum: Also, some browser tabs in desktop Firefox begin autoplaying (even though they’d been previously set to paused) after I wake my Windows 11 PC from sleep.
Offtopic question: When did LW introduce Notion-style toggles (the ones used to hide the song lyrics), and how can I use them myself? I didn’t find the answer in lesswrong.com/editor.
It’s just the native
<details>HTML element. We don’t currently support it in the editor, but just manually added it to the HTML of this post.I do think it would be nice to support it in the WYSIWYG editor, and it’s been sitting on my “to build” list for a while, but we haven’t gotten around to it.
If you do get around to implementing such toggles / HTML <details> elements in the WYSIWYG editor, I recommend checking out how Notion implements their toggles, and especially their toggle headings.
The song playlist is not in the same order as the list of songs in this post. Is there a canonical order?
Oops, that’s an error. Canonical order is the one in the music player. Will update the post later today.
Here’s another tiny Windows Firefox bug report.
Expand any song.
Hit play.
Collapse the song.
The song keeps playing. Would’ve expected the music to stop when the video player wasn’t visible.
I hate that I actually liked Answer to Job
No play button on mobile—does someone have the link to the album?
Click on the “Listen Now” button on the frontpage banner and the audio player should re-appear.
Its not showing on desktop now either
I need them on Spotify, the Litany of Tarrrrski is my favorite
Is it possible to crowdfund the necessary expenses to get this onto Spotify? Are there any significant potential licensing issues?
No huge licensing issues, I think. My guess is these should go live on Spotify within the next few days or so, we are currently waiting on their review to complete.
Today is Easter Monday, which is a holiday in the UK. I wanted to spend this morning reading Joe Carlsmith’s Otherness and Control in the Age of AGI sequence. The playlist banner taking up space at the bottom of at the bottom of my Kindle Fire screen is annoying, especially because it doesn’t quite hide the text behind it, just makes it very faint and blurry. I have spent a lot of time trying to figure out how to get rid of it and now I’m angry, not relaxed. I am not amused. Just tell me how to get rid of it, please.
Try here or here or here.
Great, thank you :-)