An odd discovery on Spotify
robinsloan.com
robinsloan.com
People will set up fake Spotify artist accounts with stolen identities and bank accounts, pay a musician for songs that pass as music, and then bot millions of streams on them. At this point there are so many of these fake profiles and songs that the music, which is simple “mood music” normally (which happens to be easy to make), is appearing on real playlists and being recommended to real listeners.
Happens to me a lot on Youtube too - you watch just one thing - and now your recommendations are full of things like that one thing. Or Facebook/Youtube, somehow: I watched a single video on Youtube about Viking sword fighting and suddenly my people recommendations on Facebook were at least 20% militaria fans, always visible from the profile picture already. It finally stopped but it took well over half a year.
That's why I thought this story was about recommendation systems recommending either only narrowly what you already know, and finding something new is not really well supported or not better than random, or that one outlier can skew your recommendations for a long time. Worst is there is no way to tell the system "stop recommending me this kind of stuff" at least for the second problem, no manual way to make adjustments for the user.
Spotify's recommendation system is much better than youtube's (at least for me). I frequently "joke" that it is one of the few applications of ML that actively make my life as a consumer better.
Having been following Spotify's "Discover Weekly" for several years now, I'm actually really impressed how it manages to blend my long-term taste with recent moods. If I've been listening to one type of music for 1 or 2 weeks, there will be a noticeable uptick of it in the recommendations, while still mixing in less recent tastes.
I've had the opposite experience with Spotify - I'd say my discovery of new music has withered to almost nil since switching when Google Music shifted to YouTube.
The algorithm just churns stuff I've already listened to, or suggests artists with (consistently) two songs that feel like Spotify's own fake artists - session musicians generated to capture royalties in-house.
Anything suggested on the front page is either songs i already have in my liked songs or completely out of place. And the songs that i've already liked are from a few specific genres and artists, which seem completely random.
A few examples:
* Ratatat and Röyksopp both are in my top 5 of all time according to Spotify's own stats, yet I never got any suggestion about E.VAX, Kunzite, or Röyksopp's releases (their Lost Tapes playlist and their latest album, released about 4 months ago).
* the 'recommended for today' section is regularly filled with random synthwave when it's a genre i barely ever listen to, and a lot of 'electronica/trance/organica/deep house/whatever you want to call it', which i don't really listen to either (at least this kind of electro music).
* still, the worst offender has to be podcasts. I have zero interest in podcasts, have never clicked on any and likely never will, yet it's always the first thing on the app homepage, just below 'recently played'.
I've tried to use the browsing categories but these are most of the time just as poor, they only contain a few playlists and the latest big releases of the genre.
Highly subjective, but i'm tired of personalized suggestions & feeds. Just because I watched or listened to something does not mean I want more of it. And imho 'just use a no history session' or 'click not interested' do not solve the problem, especially since the argument in favor of recommendations seems to be that it is 'more convenient for the user'.
Why can't I just browse instead? Spotify has a very extensive way of categorizing songs based on multiple characteristics (https://developer.spotify.com/documentation/web-api/referenc...), so why can't I just use that when searching for new music, directly in the normal app?
I hear this all the time but Spotify just plays stuff I've listened to. It's not a discovery service for me, it just plays the hits.
I'm wondering if that's just what it is and everyone likes it because it's playing stuff they already like for twelve tracks and then one new song.
I've heard reports of this happening but for well over a year, YouTube was still recommending relatively high quality videos to me that went along with my interests.
Until last week.
Somehow, somewhere along the line I must have clicked on some sort of "influencer" video. Now all I see are hundreds of hyper Minecraft videos, "I filled my house with 1 million packing peanuts" videos, etc. I have tried manually searching for some of the topics that interest me again, but these influencer-type videos still overrule whatever I try to manually teach it.
I don't know what happened, but it has ruined it for me.
I can vaguely see why YouTube is pushing right-wing conspiracy videos on me, but Minecraft? Mr. Beast? I’ve blocked perhaps two or three dozen of these channels. Take the hint, YouTube.
I put the word independent in air quotes because they all have near-identical looking title cards with the similar fonts, some variant of the phrasing "The collapse has begun!", and the same-ish content. Different presenters, different channels, same message. Over and over. And over.
Reminds me of the "This is Extremely Dangerous to Our Democracy" clip that edited together dozens of apparently independent local news channels saying the same script, verbatim.[2]
You've got to wonder if there's someone with deep pockets trying to influence public opinion on China? Or is it just the same thing as the "local" news channels, where a bunch of YouTubers have been "bought" by a corporation that send them material to read?
I would love to know if anyone on Hacker News knows something about this kind of thing...
[1] In case you're interested: https://www.youtube.com/results?search_query=stoic+finance
...and some copy-cats clips I recommend opening in a private tab unless you want to be inundated with even more clones:
https://www.youtube.com/watch?v=6slQLbT_fNY
https://www.youtube.com/watch?v=0-tLenP5NA4
https://www.youtube.com/watch?v=iDVNag9Pq7s
https://www.youtube.com/watch?v=4JEdz1eA2vQ
[2] Entertaining and terrifying: https://www.youtube.com/watch?v=ZggCipbiHwE
Either China has to indefinitely grow at 10%+ per year (it's becoming obvious that's not going to happen), or the Yuan or Chinese home prices (~50% of Chinese wealth) are massively overvalued.
These are the least likely options.
The most likely case is that the channels are all copying each other. Even if they were all put out by a couple content farms, it's extremely unlikely that the content farms care about anything else but making money. The only thing they likely care about is that people watch videos like those.
People trying to make easy money is, by far, more common on YouTube.
The saying "power corrupts" applies to this circumstance perfectly.
Spotify manipulates recommendations to what makes them the most money, not necessarily what is best for user satisfaction. They do it just enough to prevent users from cancelling their subscriptions as well (usually).
>That's why I thought this story was about recommendation systems recommending either only narrowly what you already know, and finding something new is not really well supported or not better than random, or that one outlier can skew your recommendations for a long time.
This kind of problem is exactly why we are building alternative YouTube recommendations. Search a channel to find more channels making similar content:
Click on icons in the lists to go to that channel's list. You can "surf" along that way to discover new content.
One can only handle so much Cotton Eye Joe and Black Betty...
With respect to news, these systems are driving polarization in politics.
Deanie Richardson
Danny Richardson
Dani Richardson
So, they are typo-squatting as part of the scheme.
I asssume money laundering would work by setting up Spotify accounts with stolen money (that they used to buy gift cards?), then streaming the fake songs.
Does spotify pay artists differently for paid vs free accounts? If not, then presumably they use free accounts for the bots, so it isn't really money laundering.
I guess if you get some streams from real people, then using paid bot accounts might still be a reasonably cash efficient form of money laundering.
Likely they use a front service such as TuneCore which gives them access to all the streaming services. Just using Spotify is unnecessary risky if the goal is simply money laundering.
(However, if the goal is to do pick up some royalties from all those background plays, and I think that's more common, then Spotify is clearly better)
For an example of spam/fraud, look up the album "Angry man" on Deezer. Or rather, the albums. There are 200+ of them, all with the same stock image album art. That spammer is easily recognizable, he's been doing a similar bulk upload several times per month for maybe 10 years now. He's an example of a spammer who is present on very many streaming services. Spotify has kicked him off, though.
Also, how do they succeed in keeping this scheme for more than a few days after the identity theft is noticed and reported to the police?
And why would they not just take the very music they just bought and paid for, and publish that as any other artist do? There's no crime in buying the rights from another artist -- that's exactly what all music labels do.
Because if you have a stolen identity and open a bank account, as long as that account isn't trying to defraud the bank (e.g. write bad checks, commit ACH fraud), it can go on living for a long time before it's discovered and closed.
In this case, they opened up a bank account with stolen identity, and then used that to get their Spotify payouts, and then eventually transferred that to another account (cash, crypto, whatever). From that bank's perspective, unless they were specifically informed of the bad account, there wouldn't be a ton of unusual activity that would flag the account.
Why would a stolen identity not being used to steal from the person whose identity was stolen, but just to process legal payments, be noticed in a few days? Or even a few years? Maybe at some point the IRS might find the account that used my identity and send me an angry letter, but that seems like it would be years down the line.
https://www.youtube.com/watch?v=pCAPll9A5F8
After years of grinding to break ground on spotify and seeing the names they promote taking the lion share of revenue it becomes very apparent that it's a cash grab that just can't be trusted. There are literally dozens of other sites that do the exact same schemes and only pay out a fraction of the revenue they rake in for doing little to nothing for artist and music discovery.
The social Internet just seems more scammy than ever, and that's why I keep most of my music work offline until it gets worked into a project. Eventually there will be a correction hopefully, but the vast amount of free/unrewarded work that is being hustled out of people from these sites is not sustainable, and eventually it will turn into a content creator drought/strike if it goes unchecked too long.
What you'd want to do is use the dirty money for the bots that drive up the revenue, and the clean account for the artist to collect the royalties. This way the artist has plausible deniability, and if the bank account paying for the bots gets busted, it doesn't matter, you can get a new one.
Back in the days, it was done with SMS payments and "party lines" with per-minute payments.
Buy a Spotify gift card for $10 USD cash, use that to get Spotify Premium (first use the three months free of course).
Play your (artists') songs, 30 seconds each (considered one stream by Spotify). That's 2880 streams per hour, 86400 streams per month.
Spotify pays around $2-$4 USD per 1000 streams, so that's ~170-340 USD clean money coming in. Some of that goes to Amuse/TuneCore/etc. Then music production, admin, etc. But there seems to be quite some money to be made, especially once you manage to fake your way into recommendations and get legit streams as well.
Spotify doesn’t really get your money at any point. The bots don’t need to be Spotify Premium.
And of course it would make it very difficult for legitimate businesses to survive: they would lack the implicit subsidy that the illegitimate businesses get.
Uber had (still has?) similar challenge where fake drivers were created, who took fake rides to liquidate stolen credit cards.
Makes you wonder what other media out there might be laundering schemes, too...
Of course Hollywood accounting is already "This movie took $100 million at the box office, we spent $98 million on 'marketing', using our own internal marketing company.".
I guess making crap Hollywood movies loses you money rather than laundering it...
The movie doesn't need to be crap to be a product of money laundering! The Wolf of Wall Street[1] was infamously funded by money laundering (from the Malaysian Sovereign Wealth Fund),
[1]:https://en.wikipedia.org/wiki/The_Wolf_of_Wall_Street_(2013_...
See this older piece by Natasha Lomas on Tech Crunch:
I watched 1,000 hours of YouTube Kids’ content and this is what happened… http://tcrn.ch/2iPXpIA
Anywhoo I was surprised to find close to 20 Youtube accounts that only has like stitched-together 1 hour videos of all her content. With non a none negligible view count.
I’m not saying that it never works - and I’m not saying that this was not happening in the past (in fact, it almost certainly was) but the chances of it happening today are really pretty low.
Here’s why - and I’ll also talk about what is more likely to be the reason for these tracks.
1. Operational Logistics
Although streaming activity is reported in real time Spotify accounts usage to labels and distributors in the middle of the month following the month in question. The label/distributor then invoices and receives payment. Depending on the deal the label or distributor has with Spotify they will be paid on a defined cycle. This may well be NET 30 EOM - meaning that when the distributor raises the invoice it will be paid 30 days after the end of the month in which it was raised.
2. Technical logistics
Both Spotify and distributors identify and flag unusual streaming activity. If you bot 1 million fraudulent streams this activity may well be removed in between the streams happening and payment being made. If a distributor sees sudden spikes in activity they may well hold payment until they can identify whether or not the streams are genuine.
Streaming fraud is a hot topic - and it’s obviously fairly easy to identify “non-human” activity when you have end to end control of the streaming platform.
Spotify has historically significantly reduced play count on tracks they think are botted: today I’d be surprised if many botted streams are even counted against tracks.
3. Economics
Essentially this is an arbitrage situation - can you get enough streams to drive more revenue than the amount you are paying for the fake streams?
The answer is almost certainly no.
Let’s say you’ve got 1000 accounts that are going to be your bot farm. There’s a lot of work is going to have to go into getting those accounts to stream in a human-like way.
Because Spotify does not have a user centric royalty model (where the subscription payments of a user go proportionately to the artists they stream) there is no fixed “amount per stream” - only broad indicators.
Streams from premium accounts pay a far higher rate than streams from ad supported accounts.
Roughly speaking a million premium streams is going to bring you somewhere in the region of $4000-5000 dollars.
If your bot farm is 1000 premium accounts then that’s costing you $10,000 a month. You’re going to have to have a certain amount of hardware - or hardware emulation - as if the streams on a certain track are all coming from web, and from the same network, that’s much more likely to flag than streams distributed across web, Spotify desktop app and Spotify mobile.
Now, sure, you can pay someone who already has the bot farm - but there are then three risks: first of all, that they take your money and run; second, that their bot farm gets flagged and the streams don’t pay through; third that Spotify doesn’t pay out on the streams.
So assuming that you’ve set up your own bot farm investing in 500 cheap android phones and doing the rest with emulation there’s a lot of cost - at least $10,000 a month.
Can you drive 5m streams a month? Well, each node in your bot farm would need to stream the track 5000 times a month. Is that going to look strange? For sure.
There’s also the question of finding distributors who are going to pay out easily and quickly if you can even convert the fraudulent streams to cash.
What is actually happening here? Well, mood music - ambient music, ambient electronic - is a really important part of the “make music happen” sell of Spotify. A lot of people don’t want to hear the hits. They want music to play while they work or run or cook or have a shower. And for these kinds of “activity driven” or “event driven” music it’s important that the flow of a curated playlist works.
You might have ten great tracks by known artists, but the transition between those tracks doesn’t work. Spotify has (according to media) begun commissioning songwriters to write these “filler” or “transitional” tracks. There was a huge tabloid storm whipped up about this by a music journalist some time back who said he’d unveiled a whole stable of “fake artists”. Spotify wants people to keep listening to music. I suspect - but have no direct insight into this - that people who listen to a lot of ambient playlists are particularly valuable subscribers - they will almost certainly be paying (as you don’t want your flow interrupted by ads) and they will almost certainly be over indexing against usage. They are probably also less likely to churn because of the lock in of saved playlists etc. It’s also quite likely that a lot of these custom transitional tracks are not available on other platforms so you may not even be able to replicate a saved playlist.
Of course GP might just be a GPT3 experiment...
You’d think, she shouldn’t be releasing any new works, you know, on account of being long-deceased and all. Therefore Spotify could block releases claiming her as an artist. Unfortunately, there are new renditions of her work being performed, recorded and published regularly, by many different groups. All of them have the right (perhaps even an ethical duty) to put Fanny Mendelssohn as one of the artists. These are works I’d like to hear, so Spotify’s recommendation algorithms are on the right track.
How on earth can Spotify distinguish real Fanny from fake Fanny?
I can't tell if the post is serious or just a setup now.
Sometimes it's obvious (e.g. techno DJ and death metal band with same name), but sometimes it's not even easy for a human to decide (two unrelated indie bands from the early '10s with the same name).
So it's not too much of a stretch to imagine that it can be exploited for money laundering and/or algorithm revenue farming.
In retrospect, I think it's more likely that Fake Fanny Mendlelssohn did NOT inform Spotify that they were a different person, and Spotify only split it out after a lot of Felix/Fanny fans clicked "I did NOT like this" when the track was put in their Release Radar.
posthumous releases are a thing
Classical music categorization (where the composer of the music is often of more interest than the performer) is very difficult to reconcile with pop music (where the opposite is more likely to be true).
Composer / performer is a whole nother metadata thing that should happen.
The results vary from incomprehensible to some kind of uncanny valley “is this just…really terrible writing or am I having a brain problem?” territory.
I’ve actually encountered lively discussions of these sorts of cloned articles with nary a mention of the weird writing.
With what we’ve seen from AI text and image generators recently, it’s only going to get weirder soon.
It's always giving me false hopes that my problem got resolved until I notice that I've alrezdy read the question before.
I’m glad it’s not just me! But for me, it’s not just random searches. It also happens with news articles from major outlets. Can they really not afford to pay a single person to just proof read the story and make sure whatever’s auto-generated is at least coherent?
What's fascinating about natural language models for me, is that I assume it is already being applied to "spin" articles, and I imagine that is part of why Google kinda sucks to use right now.
It's like they're basically using a fancy "arpeggiator" to generate these songs.
What's particularly interesting in the music sphere is that there are already well-established trends towards building a sort of ambient, atmospheric, generated soundscape. (For example, the famous "Lo-Fi girl" stream.) AI-generated content is a very natural progression here.
Regarding the broader pop music industry, a "GPT-3 for music" would likely further inequalize the relative power of labels and musicians. If people who control music distribution can easily make hit songs without needing to hire songwriters, arrangers, or performers, they surely will do so. I can imagine a lot of music-related occupations potentially having to pivot to rely much more heavily on live performances to make any money.
I could write a ten page rant on why AI and automation aren't a threat to musicians but it basically boils down to the fact that music is a human spectacle and we will continue to grow the industry through performing live music much like a robot that can mimic Tom Brady isn't a threat to teams selling tickets to see Tom Brady play live and people to watch it on TV (which exists, by the way).
At the end of the day, technology that lowers the barrier to entry for records and distribution is a massive boon to the industry at all levels. The rising tide lifts all boats. The companies doing interesting things with AI in music aren't just generating old shit, they're making tools to give to the next generation of creators to create new shit that no training can replicate, because it has never been done.
Some labels might try to present artists that have AI music, and no doubt it would still be consumed. But it's a huge risk when it to the human aspect of it as people want authenticity.
Music is consumed in a huge array of contexts though, where I think AI music will end up is as music in movies, backing tracks in adverts and youtube videos, as filler music for all kinds of other media.
You're really not thinking it through if you think any genre isn't going to fundamentally change once the industry starts to push virtual artists. They can be perfect and relatable to teenagers. You really don't need physical people to pull off a good festival, a well orchestrated 3d avatar is likely even better because they can be bigger and seen from the back
https://www.theguardian.com/technology/2010/jul/11/david-cop...
It's a great read! https://www.goodreads.com/book/show/719900.Little_Heroes?ref...
It's weird. You could extrude endless amounts of The Caretaker, "Everywhere At The End Of Time" if you specified the recipe, with a 'deterioration dial' and some coding to determine how you vary the output. But you'd be free-riding on a pre-existing artistic intention that was originally implemented in dramatic, bold manner. To extrude endless amounts of this stuff is both fairly trivial and missing the point completely…
Don't order food out of a random dark kitchen. Order from a brick and mortar restaurant you know.
Don't use algorithmic feeds to discover video. Find a channel/producer you like and watch that.
The same applies to music. Assume everything fed to you is a tiktok style snippet of short term gratification that has no real value other than keeping you around for a few more seconds until the next snippet, while costing as little as possible for both producer and delivery platform.
Don't waste time being upset that everything is TikTok. If you scroll the TikTok feed, Watch Instagram reels, Order Dark Kitchen food, or listen to generated Spotify playlists, you are part of the problem.
yes consumers create the demand, and yes corporations are simply fulfilling that demand, but no, changing consumer behaviour is not the solution
it is unpleasant to say, but most (all?) people are not rational actors, and no matter how much proselytising to the choir on obscure message boards goes on, that is not going to change
given the free opportunity, especially with the urging of self-interested corporations, most people will become addicted to cigarettes, or slot machines, or micro-transactions, or heroin. the solution is not to teach some bible of consumer behaviour to a group of people that probably don’t do these things anyway, it is to use public organisation to force these large, perversely-incentivised groups of people to follow a set of rules that maintain a healthy media space
I bet many listeners believe the Spotify-presented playlists are being curated, as if the stock image models in the thumbnail actually are jamming together and meeting with musicians face-to-face. So much of the recommendation service seems obfuscated to customers that I honestly don't understand what the end game is. Everyone is racing to just be "stickier" than their opponent that I can imagine a sudden pivot in customer demands where "Made for you" becomes a technology mistake akin to "Facebook Games"
Eclectic music curated by real DJ’s, you’re welcome.
i only listen to live radio, this is my station rotation (in addition to KEXP)
Which in turn already fails at the majority of people being too lazy to type, and not knowing how to search for things they want. They quickly give up, and just take the easy route of consuming what’s served to them.
It’s the modern equivalent of falling on the sofa, switching on the TV, and just watching whatever’s on until they fall asleep.
Getting a majority of people to care about their media consumption is probably a hopeless cause.
Music Business Worldwide’s story explains how the moneymaking trick worked. “Soulful Music” had 467 songs by virtually unknown artists — which is to say, artists who may have been created for the purpose of this alleged scheme. The vast majority of songs were about 30 seconds long, which is the minimum length a song needs to be to count as a monetized play on the service.
The most probabl explanation for all this is that someone or someones in Bulgaria set up 1,200 computers with premium Spotify accounts, then had them play the songs on “Soulful Music” constantly. While it would cost $12,000 to set up all those accounts, the payoff would be worth it.
Link -> https://www.inverse.com/article/41573-spotify-bulgarian-play...
Spotify pays PER STREAM, meaning that a bot account can funnel a massively disproportionate amount of revenue compared to a human one. In fact, you can generate more revenue than your subscription costs. Artists have complained about this forever, because 24/7 playlists at the gym with Justin Bieber on repeat would "steal" from the indie enthusiasts. And they are right! If I use my account to listen to one band only, my $10 contribution should go to them (modulo taxes, margins etc).
I'm sure they have "patched" some of these holes in recent times (ie some half baked abuse detection system that scammers can circumvent easily). But the per-stream principle remains, and it's such a massive incentive fuck-up from every angle, beyond salvation. And now they have (predictably) content farms and money laundering at their necks, and they're still not patching it.
I'd argue that pooling the funds and divvying it up evenly across all played songs is actually the most fair. If anything, your implementation would be more prone to scammy content farmers (e.g. spammers using trials/locale pricing arbitrage to play only one $SCAM_ARTIST song per account).
Most importantly, however, your model is just not how memberships work in the real world. HBO doesn't (and shouldn't) reward a TV show with more budget because you only binged one TV show despite having access to all the show. Nor do gym memberships. Do you think Costco shouldn't allocate your membership funds for their gas stations because you have an EV? What about cable TV? What about your local/state/federal taxes?
Wat? You're saying that if you misimplement it and count all users including free/trial as equal paying you could get problematic incentives? Well, yeah. The point is today you have correctly implemented per-stream and there is a massive open arbitrage. A bot account can control 100-1000x of revenue compared to a human.
> Most importantly, however, your model is just not how memberships work in the real world. [...]
Spotify doesn't produce music, they act as middle man between you and the artists you listen to. There's a reasonable expectation that Spotify takes a cut between you and the artists you listen to, similar to an app store. Splitting your contribution between the artists you're listening to makes sense and importantly does not have arbitrage for content farm spammers (modulo a tiny bit of value from stolen accounts).
HBO and Costco aren't just funneling money around, they're producing physical goods and content, so it doesn't apply directly. You could make a more convoluted argument that they should follow the same principles but it's really not clear what that would mean, so I'm not gonna defend it. Same with taxes, I don't know what either per-stream or per-customer would even map to in that space.
You, who listen to IndieBand™ for a total of an hour a week or so.
The Gym™ who plays Bieber 24/7 every day, all month.
You each pay $10 a month, so $20 total. Spotify allocates based on plays, so Bieber gets 720 "played hours" and IndieBand™ gets 4. So Bieber gets 720/724 of the $20.
The OP is saying in this case, Bieber should get $10 and IndieBand™ $10 (minus fees, etc).
- A Bulgarian scheme scammed Spotify for $1 million—without breaking a single law
https://qz.com/1212330/a-bulgarian-scheme-scammed-spotify-fo...
- The Rise Of Spotify Streaming Farms: How Fraudsters Are Cashing In
https://lunio.ai/blog/ad-fraud/spotify-streaming-farms/
- When Spotify, Pandora, and iTunes Pay Royalties, This Is Where the Money Goes
As Spotify's recommendations become more ingrained in how people listen to music they'll have to think about how to treat "song spam"
https://blog.echonest.com/post/48943428838/how-we-cope-with-...
Spotify is actually half-decent here. The other streaming services seem to mainly compete on cutting costs, and many are little more than frontends over 7digital's services (who, surprise, also do virtually nothing to prevent spam).
To take an example:
https://www.deezer.com/search/%22From%20the%20box%22/album
That particular spammer has been doing this thing for years. He's easily recognizable: one characteristic is the way he uses colour filters. If you search qobuz (one of these terrible front-end streaming services which does no QQ) for the label "piano to go", and say, the album "Joker Games", you'll notice it's uploaded once for Bill Haley, and once for Bill Haley & his Comets. That was apparently close enough to be detected as a duplicate, so he put a blue image filter on the latter.
Most of this guy's albums eventually get deleted (or delisted), even on sites like Qobuz - probably more for lack of plays than for being detected as spam, I'd bet. So you don't get his full history. But I found out the French library service made some effort to catalogue streaming releases a while back, and with access to search in those, you can track his evolution. He was a lot sloppier in the past, for instance he didn't bother to come up with a new label name for every dump. But he's been at it since at least 2015.
Tidal still has him, though: https://tidal.com/browse/album/241125359
E.g. in they key of C I (C E G), iii (E G B), vi (A C E). iii can be seen as Imaj7 without the root (C) and vi as Iadd6 without the 5th (G).
So basically this is I-V-I-I with some decoration, being V-I the quintessential movement in western classical music.
I guess it could help shift some streams to indie artists with a less favourable contract compared to the big 4 record companies, but it hardly seems worthwhile?
I’m gonna look into this more. I have friends trying to break into the industry and gaming the Spotify and TikTok algorithms is (unfortunately) a big part of that these days. Having fake, phantom artists filling up your flagship playlists is so anti-creator I’m truly appalled at Spotify.
https://open.spotify.com/artist/2wdHPx6lvGu3MvTH61uvTi?si=3n...
Money is too valuable and rewarding that many people will constantly chase these hustle practices no matter the ethical implications or if they are shady. Even more so if they are legal.
With the current potential of AI and automation this just makes the resulting impact bigger for the 'good citizen consumer', like the sheet amount of autogenerated videos on YouTube that just read out loud some scrapped content, or even the stolen and reposted content, all in seek of finding a semiautomatic money cow that they can milk. After all, if that can give you easy scalable money, why not trying?
The bad part is that now consumers have to deal with a polluted search result and recommendations list, that is really hard to filter by the media owners because its a tough issue.
I understand the motivations, but I hate that the product degrades the quality as a result of a few actors .
Sometimes I wonder if this will always be the status quo in our world and I'm actually missing out on joining these practices as well and see if I can hit some jackpot, but I hate these side hustle practices.
I find this infuriating. There's one dominican restaurant in my city that's listed on DoorDash, Grubhub and UberEats under over 20 different names with different subsets of their menu. All of these duplicate listings for the same place make it harder to search or browse the actual restaurant options in the area. Any time I come across a restaurant that looks worth trying out, I have to look up the address to see if it actually exists, or if it's just IHOP masquerading as a burger bar.
That would actually provide a service that doesn't dilute their brand. As it is, these restaurants have just found a way to counter this negative outcome by diluting this third party marketplace itself.
What's wrong with food from a ghost kitchen?
There's no Saladworks if you visit their address, just the pizza place. They offer the same menu as Saladworks, the chain of fast service salad shops, but don't have the same supplier or training or equipment as a Saladworks. This pizza place just attempts to recreate the same salads using inferior ingredients, pizza toppings, generic dressings, and generic plastic takeout containers. The result doesn't taste like a Saladworks salad and is smaller than a Saladworks salad. I've tried it twice and neither recreation of Saladworks was faithful, missing ingredients both times, with a different lettuce mix than Saladworks uses, etc.
There's no Guy Fieri restaurant if you visit their address either. The food from this brand of the ghost kitchen is just inedible. It has 5 reviews on Yelp locally so far, and all are one star. They're microwaving burgers, sending out raw pasta, sending out easy mac with pizza toppings as a fancy loaded mac & cheese, using the same generic plastic takeout containers they do for Saladworks.
And yet at least they have a brand to burn. What I was actually frustrated with in my comment a few levels up is the places that list their own menu, sliced and diced in different ways, under 20+ different names. If someone complains about the quality of "Simply Crepes", they can just remove that listing, since those crepes are also available under "Morning Times Cafe", "Morning Breakfast Sandwich Bar", "The Daily Fare", "The Coffee Creperey", etc.
Here's a commercial ghost kitchen explanation to restaurateurs:
https://cloudkitchens.com/blog/ultimate-guide-to-ghost-kitch...
And some of the other ghost kitchens / home address food prep is pretty good. Like there was an ice cream place that delivered until midnight ( out of somebodies garage ) their ice cream was cooled down to well below zero so it was not melted when you got it.
How? If a choice of "this restaurant" or "that restaurant" both take you to "this restaurant", is that really a choice?
I remember the first time I got caught by this, I excitedly ordered from a "Korean fried chicken" place on Uber eats, and when the food arrived it was heat-from-frozen chicken tenders drowned in unappetizing sauce. I felt scammed, and the food was so bad, I didn't even manage to finish what I had ordered. Later I looked up the restaurant and discovered it shared an address with an unremarkable local burger chain. I had a similar experience with another virtual restaurant and at this point I always check if it's a VR before ordering delivery from an app and treat it as a strong negative signal to the quality of food.
This is obviously a lot different from your hotel example because the hotel restaurant's reputation is inextricable from the hotel itself; not only do patrons always know where their room service is coming from, bad food/service reflects poorly on the hotel as well. The hotel has a lot more to lose than the virtual restaurant.
The multitudes of different places at Disneyworld that all operate out of the same massive underground/hidden kitchen complex might be a better example.
Yes. Hotels with multiple restaurants tend to have only one kitchen, because why wouldn't they?
Again, I don't understand what's so offensive or controversial about this concept. I'm probably misunderstanding the objection.
His 2 novels, "Mr. Penumbra's 24 Hour Bookstore" and "Sourdough", are both really engaging reads. His 3rd novel is in progress and it seems like it's going to be more of the same quality. He also has a couple of novellas and a bunch of great short stories (some of which he did limited printings of and mailed out and one of which was procedurally generated and unique to the person who purchased it).
His newsletter is constantly filled with great media recommendations and he's always working on something interesting in that bridges the media and tech worlds.
He also makes great olive oil (Fat Gold).
I am subscribed to his newsletter but I confess I rarely if ever read it. After having read this post and your comment, I'll make an effort to give it a solid look.
Festina lente!
This is almost certainly generated as well, found through a recommendation series from the article.
It's so creepy because it feels like something that isn't made by a human, like composition uncanny valley!
Edit: As a side note, you should be careful to not like the music in this article. I suspect it will poison your Discover Weekly and other algorithmically generated playlists as they seem to be based on other listeners' data.
Here are some other interesting seeds.
* 0:47 seed (Romilda Prime): https://open.spotify.com/playlist/37i9dQZF1E8NyZRPJ6i5rq?si=...
* Venera Fanucci seed: https://open.spotify.com/playlist/37i9dQZF1E8PTURLXiSzw3?si=...
* Scars Hayden seed: https://open.spotify.com/playlist/37i9dQZF1E8NLdq3OfZZrl?si=...
* Chinpe seed: https://open.spotify.com/playlist/37i9dQZF1E8O73Ntg828PD?si=...
This one is interesting because even the album art is generated, rather than stock. The artist names are usually all uppercase. Perhaps a different author, but the techniques seem the same.
* Western Wilds seed: https://open.spotify.com/playlist/37i9dQZF1E8LLKjmBVSRwb?si=...
(See also Exboro Key, Defiant Leather, Gallisle Isle...)
* Hollow Linen seed: https://open.spotify.com/playlist/37i9dQZF1E8PeWgO6Bpowe?si=...
* Surrane Path seed: https://open.spotify.com/playlist/37i9dQZF1E8NCJ0IQOCfKi?si=...
This is a broken classical music generator, perhaps the most interesting!
Rdio used to have a top played section which for a while got dominated by some very very weird albums. This was one of them:
https://open.spotify.com/track/0e3DihEy2kupmSSpzqfqNP
The story behind how it got to the top is outlined here:
https://musicfeeds.com.au/news/melbourne-programmer-hacks-hi...
Is the only solution to retreat to campfire songs and home cooking?
Wonder what the thought process was behind this. Less likely to skip the track while trying to find the groove? More monetization with the short length?
https://open.spotify.com/playlist/2IaWgbhpPbS3Z9DYgf1rqg?si=...
If someone managed to generate a collection of songs that are all similar in the right ways to pop up when people leave the recommended playlists playing, they could rack of thousands of plays without all the trouble of having musical talent and writing good songs.
https://www.chosic.com/spotify-playlist-analyzer/?plid=2IaWg...
Interesting to notice that even when there are only little variations on the same song, the analysis (which is powered by the Spotify API) shows very different indicators for each of the songs in danceability, accousticness, valence, etc. This makes rather unexpected that all those songs could come together in a recomendation system.
This is all definitely generated algorithmically using a really common chord progression in western music. (Think Pachebel's canon).
So a person (or group of people?) is creating generatively created music that is designed to be as pleasing to the ear of a human so that Spotify ranks it higher in automatically generated playlists which means more plays, which means more money.
It's crappy and spammy but also really kinda clever.
"I love it, because it’s so strange, so dizzying, and — credit where due–because our mystery producer is truly going with the grain of the medium, in a way that no one merely “making albums” does, at all. What could be more 21st century, more “liquid modernity”, than releasing your music as a haze of variations into the swirling currents of the algorithm?"
But tons of artists have done the same in the past, whether in streaming platforms or in the vinyl and CD era. The most famous is probably Brian Eno who has made several albums using algorithmic generation.
In general, algorithmic composition is a sizable niche, both in academy (tons of music PhDs on that), live performance, and recording settings. And yes, people release those things too.
Even "generated content" to make bucks is a well known thing, both in music platforms like Spotify, in Amazon (tons of auto-generated Kindle books), and in YouTube (https://medium.com/@jamesbridle/something-is-wrong-on-the-in...).
The strange thing is, the author should already have known all this, as his previous post goes:
"My band The Cotton Modules, formed with the composer Jesse Solomon Clark, also goes with the grain of the 21st century: our process combines AI tools with human skill and imagination, metabolizing a huge archive of recorded music into something genuinely new and exciting."
These generated pieces of text, art, music, etc. are mere mashup copies without sense and purpose. None of the tricks of trade and conventions the artists know and apply to their art (such as, for pop music, a standard pop song is approx. 3 to 4 minutes long, has a verse, a chorus, a bridge which are played and repeated in a specific sequence) are recognizable here. Even the deviations from said conventions are so obviously non-intentional and random, without sense or 'dramaturgy', build-up and release of tension, etc. It's not music.
If this is our future, it is only if we become less educated and, sorry, somehow significantly dumber.
And if, according to the numbers on Spotify, people willingly listen to this garbage, I only have a misanthropic "we've earned it" for you.
The only takeaway from this article is that people are staggeringly stupid and uneducated and will consume whatever trash whichever industry will throw at them.
Perhaps it's time we spent money on education again.
One of those was by an artist called Adal, and while it wasn't what I was looking for at all, I still listened to it. It was slow, rythmic and very simple music (think "human music"). When I looked for other works of this Adal, they had maybe like 8 songs, all of them the same simple melody but slightly different speed, rhythm and so on. At first I thought I was listening to that same song, but there definitely were slight differences.
At the time I didn't understand why anyone would produce such a thing, but in this context it makes much more sense.
I just searched again for Adal - Dyson Sphere, but there seem to be no hits. I wonder if the artist has since been deleted.
Spotify has collected user "skip data". That is a massive dataset. They have research to try to train AI to generate audio that humans are less likely to skip.
These results seem a bit more midi based, and I would assume that the models they generate would work on raw audio, so doesn't seem like a perfect fit.
Though, in general, it must be tempting for Spotify researchers to test their creations on their customers.
(Note: I do not believe it is ethical to collect skip data, or any data from customers and use it to train AI without a very steep OPT-IN process. Burying some legal disclaimers in wall of signup jargon is a terrible practice.)
Uggh. I wish payola were actually illegal in practice.
> So who’s really losing in this equation?
Possibly, older out of touch-artists who think music should always be something meaningful and “culturally important” and are perhaps – just slightly – butthurt that no one cares, or that other people have figured out another way to be successful.
I can't tell if this bit is dripping with sarcasm, or if the author is actually this contemptuous of musicians and this dismissive of the difficulty associated with getting "discovered" these days.
Either way, it is a fascinating article. Wouldn't have guessed that Sony was one of the leading spammers/fraudsters (see other threads on this article) in this space, or that Spotify was complicit.
This actually makes me angrier than the Sony rootkit fiasco.
Thanks for the link.
I found it totally fascinating myself. And now that I've read it I see the effects everywhere (musically).
They basically either have salaried/contracted musicians or pay a one time fee without royalties to make passable songs in every genre. Then they push their own songs in lists so people will listen to them more than "real" artists.
https://www.rollingstone.com/pro/features/fake-artists-have-...
In his book Sourdough, he spends quite a while describing the music of the Mazg - a fictional group of European migrants. For the audiobook version of Sourdough, he trains a SampleRNN model on a particular genre of Croatian music, then has the model produce Mazg music. The description of machine learning is vivid and approachable, and the audio clips are fantastic.
ie how "effective" is this, compared to alternatives? I had a belief the rate of IPR payment was low, so this is probably a low rate of return, high cost, fully digital method.
Which in my theory, would also be making spotify significant profit, as the facilitator. Are they now at some legal risk if they can "detect" this traffic?
That doesn't seem like it would be sustainable if only 0.00000001% was caught.
This scheme was so clearly a laundering play, but everyone pretended it was some amazing exit. I'm confident the tech industry is full of these shenanigans, but I've never seen anything like this uncovered.
Edit: or perhaps Bruce Sterling, as I'm sure they'd tell me.
The experimental artist Arca released an album[1] in 2020 which consisted of 100 AI-generated remixes of her song Riquiqui. It was part pretentious artsy experiment and part marketing stunt for some AI company, but an interesting concept nonetheless.
https://open.spotify.com/track/35lnhh0LCDZOu10Lasv6bi?si=257...
and lots of others that were not on spotify :
Boss Through the Author by MemethSoyal
Ariya Saunders by Raphy Fraser
So it means it is possible to catch them, and spotify is probably actively trying to remove them.
It could be that you're right, that this fake artist used to be on Spotify but was removed. But that doesn't explain why you can find e.g. Romilda Gebbia on Spotify but not on YouTube.
When I think about it, it could also be that the spammer is on multiple services, but they use different algorithmically generated artists for each service.
What was extremely useful in you finding this on YouTube, is that we can see by which distribution service they got access to the streaming services (information that all the other services hide, annoyingly). It's a French World Music thing.
Which is especially interesting, since I also traced back the "Belle Wood" spammer to a French World Music thing, although a different one (divercities/diMusic). I wasn't 100% sure if they had anything to do with the spam, they could plausibly be a victim of it themselves, but now I have more circumstantial evidence. Thanks!
(If you know of a less noisy forum where we could discuss tracking down/locking out algorithmic spammers, I would be grateful.)
I mostly listen to metal, but one day I noticed that on my last.fm my most listened to track was suddenly "The 1975 - Chocolate". A song that I'd never heard before or since. The track was played at completely random times, often multiple times in a row or when I wasn't awake.
I think that with a standard Ablenton project, for example, with several VSTs (this sounds like a Kontakt library) and replacing the midi clips by the Magenta's MusicVAE output for example, you could make infinite variations.
There's minimalistic music, and there's low effort music, but this is not that. This is spam.
Plot twist: the author of TFA also made the tracks.
Is this related to that thing where one restaurant creates several fake restaurants in food delivery apps?