I like the blog but the premise of the blog is an engineering/epistemological perspective on the craft. The writer clearly cares more about the process, technique and history more than the feeling and validation.
It could be, that a big part of the the future of hobby's and entertainment in this way is the feeling and validation over the actual performance. Or it can be that a massive amount of people find their value in this content.
The idea that we could create a world where 'a big part of the future of hobbies and entertainment' is people listening to meaningless words made up by machines that help them feel good about themselves sounds horrifying. How could anybody feel ok about that? What would it say about the society we've built?
It would say that society changes, and people who were not used to a new world get upset about it, as it has always been throughout the entire history of humanity.
We were used to having psychologists and doctors in person, now the most common form is to have it through apps, and the younger generation does not care, it's in fact more efficient to get a prescription that you like than to spend time going places and having in-person meetings. But older generation finds it hollowing out and horrifying.
You need to accept that society moves on, and it can look different from your perspective.
A looooot of assumptions here. We have yet to see any of these brave new ideas actually work.
Therapy has never been more available, yet mental health is through the basement.
I’m also not seeing any evidence that young people are the driving force behind turning the world to shit. Every Gen Z person I know craves authenticity, connection, and meaningful work. All of this is the opposite.
In this case, the user is deciding that they choose what progress is. I am saying, that people who use the tool and value the utility of it decide what is progress. If people listen to the podcast, or use doctors in the phone because it provides them any value, it will be a change and a perceived progress for them.
If the generated podcasts did not bring any value to the users, such as validation, or engagement, they would not use them, and there would be no change.
So .. I think we need to ask a deceptively simple question here, which is: is knitting real?
I'll add in an aside to this, which is not only are there fake knitting podcasts there are fake knitting and crochet patterns, which is a problem because people get a substantial way through making them only to discover that they don't work. In some cases the giveaway is that the supposed final image isn't physically possible, like the images in this article, but the fakers can use a real stolen image and just spam a pattern underneath it.
So: what is the knitting that is real? It has to be the use of your hands, needles, and yarn to produce a physical object, right?
The podcasts work towards something else. The identity of "being a knitter". This is a form of "hobby" that was already not unusual, that of discussing a thing without ever bothering to actually do it. Photographers are especially bad at this: too many lenses, not enough photographs. They've also got comprehensively run over by AI, because you can just generate the photographs now. Same for "authors".
But ultimately all these pleasant sensations aren't backed by a connection to the real. If you're going to talk about the history of knitting, shouldn't it be the real, evidenced history? As done by real (usually) women? Otherwise you're just knitting a pleasant fantasy for yourself.
The AI approach is "wireheading": the logical conclusion of all of that would be to find a means of inserting a wire in your head that provides constant pleasant sensations. Achieving happiness through a constant feed of generated images is less effective, but it's the same order of things.
(see also: authenticity in food, which could easily turn into another ten thousand words)
> But ultimately all these pleasant sensations aren't backed by a connection to the real. If you're going to talk about the history of knitting, shouldn't it be the real, evidenced history? As done by real (usually) women? Otherwise you're just knitting a pleasant fantasy for yourself.
If the real is the feeling you get from listening to the podcast or identifying with a subculture, then that is the real for that person. Factual, grounded information is just one take. If it was not this way, we would have much less myths, religions, etc historically.
People will feel the same degree of joy and completion when the final word of the podcast is read like you feel when you finish a really complex piece of work.
If you only care about the material and physical utility of the product, you can order the sweater from AliExpress for 5% of the cost and no time spent.
No, but to misinform people you have two main strategies: limiting through tailored scarcity and dilute in extra-generic overabundance. Don’t get it wrong: both can be combined and even can sometime overlap.
It doesn’t matter if no one is listening. Equally saturating all channels, metrics and indicator is enough to create hindrance so preventing relevant information to spread in meaningful time.
Attention is all you need, so distraction is all that will be given.
I wonder if (or, more accurately hope that) this kind of slop will eventually die out as people realise how little care is put into it. I am more and more convinced that if the devil existed he'd take care of the bigger stuff, but have an army of little devils that encourage people to do things like make unsupervised automated podcasts about knitting, relentlessly chipping away at the messy joys of living.
For a long time I thought that the AdSense business model was ultimately doomed because I assumed that people hate ads as much as I do. It turns out I was just wrong about what most people are willing to put up with.
I remember visiting a friend over a decade ago, and for some reason I had to use their computer for a bit. I was immediately thrown aback by all the ads everywhere and installed an ad blocker before anything else. They were very grateful, but the part that surprised me was they were annoyed by the ads but never thought to look for some way around it. It never even crossed their minds it could be done or to search for it.
At the start of Good Omens, there’s a scene where demons are sharing their recent misdeeds. A couple of them are sharing “classic” demon stuff like killing and possessing, but Crowley (the protagonist demon) shares more modern evil deeds, such as creating traffic jams.
I'm afraid it'll lead to a weird music-ification of content.
Music can make you feel good and keep you engaged just purely out of engaging our pattern recognition.
AI videos and photos seem to have a similar effect. Even if it's not real, they encode enough patterns from good human work to be able to engage our attention.
Just proving people with an attentional escape is valuable on the internet.
Yeah, people will reflexively filter out the slop, eventually, but they'll do it by leaving the places that have been rendered worthless by its persistent presence.
The particular type of innovator ghoul that's enabled by generative AI dreams of filling the entire internet with bullshit content. Aggregators (media and content) should be actively pushing them out for their own long-term survival, IMO.
Interestingly, Inception AI seem to have pivoted from content slop for "gardening, [...] knitting, cooking" - or "things we can afford to be wrong" - to "AI Immigration Drafting Software for Law Firms": https://www.inceptionai.co/
I'm somewhat curious how that'll work out. Hint: I'm not.
I remember this kind of slop from times well before the LLM explosion.
I'm specifically thinking of a print magazine that was designed to make you feel like you are a smart reader of science articles, without any useful information about the actual science or technology.
Yes, the article acknowledges this in the first paragraph by citing Harry Frankfurt’s „On Bullshit“ (1986). Of course bullshit (as well as even more insidious misinformation/propaganda) have always been around, but the incredible advances in its production and dissemination are worth considering. At some point, sheer quantity turns into its own quality. Indeed I would argue these issues have always been underconsidered. The article is a kind of inoculation against bullshit that every generation requires again and again. People aren’t born nearly skeptical enough, and the game keeps ever changing.
I actually don’t think the article is sufficiently vehement in calling out just how brain-frying this is. And how destructive on a societal level. The razor’s edge between being too uncritical and too cynical is hella narrow.
> I remember this kind of slop from times well before the LLM explosion.
Even if that were true (which I don’t think it is, this is a different kind of worthless content), you most definitely don’t remember it at this scale, and that’s a major point.
Why does this site want to access apps and services on my local network?
On topic, I do wonder how "the market" is going to sort this out. At this moment I'm leaning towards just banning this shit, but maybe there is a better way?
We can already see the market in action. Increasingly people are more hostile to online content and influencers, except for the few people they follow, just like everyone was already defensive against unsolicited email. Authenticity will become valuable in a sea of slop, and making high budget productions (think Mr Beast) will be worth nothing since it can be easily faked and hard to distinguish.
TL;DR: there are brainrot farms with help from AI.
But I saw this one coming three or four years ago.
Actually, I've been listening to AI-generated brainrot music. I prefer it to some human-generated brainrot music (there's "I Hate Boys" from Christina Aguilera. Sorry if you are a fan).
Brainrot serves a specific social purpose: relieving stress, incoherently winning elections. It's a kind of drug that dulls the dangerous part of the brain while leaving the he-is-a-good-tool and she-is-blonde brain hemispheres in working order.
In fact, I do believe that if there were to be an uprising in a couple of decades against AI, and the human side were to rise victorious, the aftermath's social order would be studiously anti-AI and anti-science, but they would make a carve-out for AI brainrot (yes, I published a short fiction story with that premise, because I'm brainrot-vers).
Are you serious when you connect anti-AI sentiment to anti-science sentiment?
To me, they are opposite sentiments, and my experience discussing AI with others supports this. The most pro-AI people I meet are very far removed from science, and my research colleagues are definitely more critical of AI than not.
ummmm, WOW!, hey that clicks
your brainrot/drug description is good.
making a choice for zero human content and therfore interaction.
the full suite of options would include perfectly artificial scents.
personaly, I am way over in the analog/organic direction, but I get the need
to disconect from the "whatever this is™" that passes for a society.
the question remains for AI scaling to meet the demands and desires society has always placed on indivuals
the audible exasperated noise comming from the person in line with me, seeing me pull out cash, thereby breaking there own perfect
little automated world, mearly by bieng subjected to witnessing such a primitive ritual, not behind me I might add, the person leaving in front of me, is the prime
example of someone who will violently reject AI and the rest when it inevitably fails to "fix" everything
Extremely long winded. I think this person is trying to throw stones at someone else’s work, but their own is so elliptical I lost the will to find out.
Not taking away the right to your opinion, but I couldn't disagree more; I found it an excellent sociological article. One, it takes the formal concept of "bullshit" and applies it to knitting in a very methodical and strict manner. I found it novel and convincing, and the examples were great; not contrived or forced at all. IMO it was much better than many academic books or articles; an immediate share.
Two, the turns of logic are clearly laid out, in a conversational way, which would make it easy to stick a wrench in and form a polemic if you found any of her arguments or logical implications specious. That said, that does make the article quite long. But then, it is anything other than "elliptical", which I think you used as "runs in circles and repeats itself often", while it actually means "omits parts and thus is difficult to understand" (like the ellipsis sign: …).
Also: what the heck is wrong with that podcast farm founder. I hope they have a bad year.
You only had to reach the second paragraph to find the example of an 8-person company that uses AI to generate “about 3000 podcast episodes per week, hosted by AI personalities.”
I was a couple of images in before I sussed it. Bullshit images, but pleasing enough to look at. Without the images, it would have either been a big wall of text, which would have put me off reading, though I did give up about 25% of the way through after sussing the images and thus the incoherence in the argument.
The images bring something to the article. They were cheap/quick to generate. The increase the potential payoff (more reader) without significantly increasing the cost. Without the images, the payoff(readers) would likely have been lower, below the cost of actually writing the article. Same goes for a history of knitting podcast or that video. Production costs would not be worth it for a very niche viewership.
Reading that made me feel like you wanted to be contrarian from the get-go and dismiss the article with the least effort possible. The whole point of the images is that they're low-effort AI slop, it's part of what she's trying to point to when someone is generating unsupervised automated podcasts about knitting.
So you're saying you can spot AI generated bullshit, but not spot a deliberate and hilarious contrivance that the author uses to reinforce their point?
I like the blog but the premise of the blog is an engineering/epistemological perspective on the craft. The writer clearly cares more about the process, technique and history more than the feeling and validation.
It could be, that a big part of the the future of hobby's and entertainment in this way is the feeling and validation over the actual performance. Or it can be that a massive amount of people find their value in this content.
The idea that we could create a world where 'a big part of the future of hobbies and entertainment' is people listening to meaningless words made up by machines that help them feel good about themselves sounds horrifying. How could anybody feel ok about that? What would it say about the society we've built?
It would say that society changes, and people who were not used to a new world get upset about it, as it has always been throughout the entire history of humanity.
We were used to having psychologists and doctors in person, now the most common form is to have it through apps, and the younger generation does not care, it's in fact more efficient to get a prescription that you like than to spend time going places and having in-person meetings. But older generation finds it hollowing out and horrifying.
You need to accept that society moves on, and it can look different from your perspective.
A looooot of assumptions here. We have yet to see any of these brave new ideas actually work.
Therapy has never been more available, yet mental health is through the basement.
I’m also not seeing any evidence that young people are the driving force behind turning the world to shit. Every Gen Z person I know craves authenticity, connection, and meaningful work. All of this is the opposite.
The problem is, who is moving “society on” and what is their agenda.
I don’t think it’s healthy to encourage an attitude to just accept all change without any sort of reflection or push back.
You could justify basically anything with that logic. Change isn't always about progress.
In this case, the user is deciding that they choose what progress is. I am saying, that people who use the tool and value the utility of it decide what is progress. If people listen to the podcast, or use doctors in the phone because it provides them any value, it will be a change and a perceived progress for them.
If the generated podcasts did not bring any value to the users, such as validation, or engagement, they would not use them, and there would be no change.
So .. I think we need to ask a deceptively simple question here, which is: is knitting real?
I'll add in an aside to this, which is not only are there fake knitting podcasts there are fake knitting and crochet patterns, which is a problem because people get a substantial way through making them only to discover that they don't work. In some cases the giveaway is that the supposed final image isn't physically possible, like the images in this article, but the fakers can use a real stolen image and just spam a pattern underneath it.
So: what is the knitting that is real? It has to be the use of your hands, needles, and yarn to produce a physical object, right?
The podcasts work towards something else. The identity of "being a knitter". This is a form of "hobby" that was already not unusual, that of discussing a thing without ever bothering to actually do it. Photographers are especially bad at this: too many lenses, not enough photographs. They've also got comprehensively run over by AI, because you can just generate the photographs now. Same for "authors".
But ultimately all these pleasant sensations aren't backed by a connection to the real. If you're going to talk about the history of knitting, shouldn't it be the real, evidenced history? As done by real (usually) women? Otherwise you're just knitting a pleasant fantasy for yourself.
The AI approach is "wireheading": the logical conclusion of all of that would be to find a means of inserting a wire in your head that provides constant pleasant sensations. Achieving happiness through a constant feed of generated images is less effective, but it's the same order of things.
(see also: authenticity in food, which could easily turn into another ten thousand words)
I am with you until you make this assumption:
> But ultimately all these pleasant sensations aren't backed by a connection to the real. If you're going to talk about the history of knitting, shouldn't it be the real, evidenced history? As done by real (usually) women? Otherwise you're just knitting a pleasant fantasy for yourself.
If the real is the feeling you get from listening to the podcast or identifying with a subculture, then that is the real for that person. Factual, grounded information is just one take. If it was not this way, we would have much less myths, religions, etc historically.
People will feel the same degree of joy and completion when the final word of the podcast is read like you feel when you finish a really complex piece of work.
Can't wear feelings and validation...
If you only care about the material and physical utility of the product, you can order the sweater from AliExpress for 5% of the cost and no time spent.
Am I to believe that those 700K+ downloads are organic traffic? Who's listening to all this stuff?
Or maybe Ms. McHealy was simply lying.
No, but to misinform people you have two main strategies: limiting through tailored scarcity and dilute in extra-generic overabundance. Don’t get it wrong: both can be combined and even can sometime overlap.
It doesn’t matter if no one is listening. Equally saturating all channels, metrics and indicator is enough to create hindrance so preventing relevant information to spread in meaningful time.
Attention is all you need, so distraction is all that will be given.
I wonder if (or, more accurately hope that) this kind of slop will eventually die out as people realise how little care is put into it. I am more and more convinced that if the devil existed he'd take care of the bigger stuff, but have an army of little devils that encourage people to do things like make unsupervised automated podcasts about knitting, relentlessly chipping away at the messy joys of living.
For a long time I thought that the AdSense business model was ultimately doomed because I assumed that people hate ads as much as I do. It turns out I was just wrong about what most people are willing to put up with.
I remember visiting a friend over a decade ago, and for some reason I had to use their computer for a bit. I was immediately thrown aback by all the ads everywhere and installed an ad blocker before anything else. They were very grateful, but the part that surprised me was they were annoyed by the ads but never thought to look for some way around it. It never even crossed their minds it could be done or to search for it.
All human progress in history has been due to a VERY small handful of people who think “this is bullshit, things could be better”.
The vast majority of people accept what they see as the way things are and it never occurs to them that things could be different.
At the start of Good Omens, there’s a scene where demons are sharing their recent misdeeds. A couple of them are sharing “classic” demon stuff like killing and possessing, but Crowley (the protagonist demon) shares more modern evil deeds, such as creating traffic jams.
https://en.wikipedia.org/wiki/Good_Omens
I’d link to a clip of it, but to your point some devil is making it frustratingly hard to find.
I'm afraid it'll lead to a weird music-ification of content.
Music can make you feel good and keep you engaged just purely out of engaging our pattern recognition.
AI videos and photos seem to have a similar effect. Even if it's not real, they encode enough patterns from good human work to be able to engage our attention.
Just proving people with an attentional escape is valuable on the internet.
It's definitely the sort of thing that Crowley from Good Omens would be working on.
Yeah, people will reflexively filter out the slop, eventually, but they'll do it by leaving the places that have been rendered worthless by its persistent presence.
The particular type of innovator ghoul that's enabled by generative AI dreams of filling the entire internet with bullshit content. Aggregators (media and content) should be actively pushing them out for their own long-term survival, IMO.
Interestingly, Inception AI seem to have pivoted from content slop for "gardening, [...] knitting, cooking" - or "things we can afford to be wrong" - to "AI Immigration Drafting Software for Law Firms": https://www.inceptionai.co/
I'm somewhat curious how that'll work out. Hint: I'm not.
EDIT: My bad, wrong company, it's "Inception Point AI": https://www.inceptionpoint.ai/
There's also https://www.inceptionlabs.ai - it's not confusing at all.
I remember this kind of slop from times well before the LLM explosion.
I'm specifically thinking of a print magazine that was designed to make you feel like you are a smart reader of science articles, without any useful information about the actual science or technology.
Yes, the article acknowledges this in the first paragraph by citing Harry Frankfurt’s „On Bullshit“ (1986). Of course bullshit (as well as even more insidious misinformation/propaganda) have always been around, but the incredible advances in its production and dissemination are worth considering. At some point, sheer quantity turns into its own quality. Indeed I would argue these issues have always been underconsidered. The article is a kind of inoculation against bullshit that every generation requires again and again. People aren’t born nearly skeptical enough, and the game keeps ever changing.
I actually don’t think the article is sufficiently vehement in calling out just how brain-frying this is. And how destructive on a societal level. The razor’s edge between being too uncritical and too cynical is hella narrow.
> I remember this kind of slop from times well before the LLM explosion.
Even if that were true (which I don’t think it is, this is a different kind of worthless content), you most definitely don’t remember it at this scale, and that’s a major point.
Why does this site want to access apps and services on my local network?
On topic, I do wonder how "the market" is going to sort this out. At this moment I'm leaning towards just banning this shit, but maybe there is a better way?
> I do wonder how "the market" is going to sort this out
Unlikely to do a better job than it did with anything else.
We can already see the market in action. Increasingly people are more hostile to online content and influencers, except for the few people they follow, just like everyone was already defensive against unsolicited email. Authenticity will become valuable in a sea of slop, and making high budget productions (think Mr Beast) will be worth nothing since it can be easily faked and hard to distinguish.
TL;DR: there are brainrot farms with help from AI.
But I saw this one coming three or four years ago.
Actually, I've been listening to AI-generated brainrot music. I prefer it to some human-generated brainrot music (there's "I Hate Boys" from Christina Aguilera. Sorry if you are a fan).
Brainrot serves a specific social purpose: relieving stress, incoherently winning elections. It's a kind of drug that dulls the dangerous part of the brain while leaving the he-is-a-good-tool and she-is-blonde brain hemispheres in working order.
In fact, I do believe that if there were to be an uprising in a couple of decades against AI, and the human side were to rise victorious, the aftermath's social order would be studiously anti-AI and anti-science, but they would make a carve-out for AI brainrot (yes, I published a short fiction story with that premise, because I'm brainrot-vers).
Are you serious when you connect anti-AI sentiment to anti-science sentiment?
To me, they are opposite sentiments, and my experience discussing AI with others supports this. The most pro-AI people I meet are very far removed from science, and my research colleagues are definitely more critical of AI than not.
AI's tendency to emit unsourced, untrue statements with authority is about the most unscientific thing you can get.
AI is scientism: presenting science-flavoured things as a cultural marker.
ummmm, WOW!, hey that clicks your brainrot/drug description is good. making a choice for zero human content and therfore interaction.
the full suite of options would include perfectly artificial scents. personaly, I am way over in the analog/organic direction, but I get the need to disconect from the "whatever this is™" that passes for a society. the question remains for AI scaling to meet the demands and desires society has always placed on indivuals
the audible exasperated noise comming from the person in line with me, seeing me pull out cash, thereby breaking there own perfect little automated world, mearly by bieng subjected to witnessing such a primitive ritual, not behind me I might add, the person leaving in front of me, is the prime example of someone who will violently reject AI and the rest when it inevitably fails to "fix" everything
Extremely long winded. I think this person is trying to throw stones at someone else’s work, but their own is so elliptical I lost the will to find out.
Not taking away the right to your opinion, but I couldn't disagree more; I found it an excellent sociological article. One, it takes the formal concept of "bullshit" and applies it to knitting in a very methodical and strict manner. I found it novel and convincing, and the examples were great; not contrived or forced at all. IMO it was much better than many academic books or articles; an immediate share.
Two, the turns of logic are clearly laid out, in a conversational way, which would make it easy to stick a wrench in and form a polemic if you found any of her arguments or logical implications specious. That said, that does make the article quite long. But then, it is anything other than "elliptical", which I think you used as "runs in circles and repeats itself often", while it actually means "omits parts and thus is difficult to understand" (like the ellipsis sign: …).
Also: what the heck is wrong with that podcast farm founder. I hope they have a bad year.
You only had to reach the second paragraph to find the example of an 8-person company that uses AI to generate “about 3000 podcast episodes per week, hosted by AI personalities.”
Yeah, well good thing that LLMs are good at summarizing articles, unlike generating believable knitting images.
I was a couple of images in before I sussed it. Bullshit images, but pleasing enough to look at. Without the images, it would have either been a big wall of text, which would have put me off reading, though I did give up about 25% of the way through after sussing the images and thus the incoherence in the argument. The images bring something to the article. They were cheap/quick to generate. The increase the potential payoff (more reader) without significantly increasing the cost. Without the images, the payoff(readers) would likely have been lower, below the cost of actually writing the article. Same goes for a history of knitting podcast or that video. Production costs would not be worth it for a very niche viewership.
Reading that made me feel like you wanted to be contrarian from the get-go and dismiss the article with the least effort possible. The whole point of the images is that they're low-effort AI slop, it's part of what she's trying to point to when someone is generating unsupervised automated podcasts about knitting.
The AI images were deliberate and part of the narrative. Ie, you can generate slop with zero effort.
from TFA: "All of the images in this post were generated by an ai in response to the simple two-word prompt “lovely knitting”
Edit: ps: Kate Davies is an actual creator who has been creating knitting patterns for years.
So you're saying you can spot AI generated bullshit, but not spot a deliberate and hilarious contrivance that the author uses to reinforce their point?