✘ From MySpace to Content ID to AI Slop
And: Is Bandcamp the answer?; Why listeners interpret songs differently; Make something heavy; Open models, closed networks; Lo-fi art, human tools
I’ve been seeing some nostalgic takes of early Web2 social platforms recently. Endless longing for the simpleness of Tumblr, for example. Or Aleena Vigoda reminiscing on how MySpace’s biggest feature - being able to personalize your page through custom HTML/CSS - was a coding error. She goes on to say that this feature was so popular it led to people creating how-to guides: “Sharing and remixing templates became a social feature—a channel for connection and engagement.” This sounds a lot like what every new create-don’t-just-consume app is trying to achieve. We’ve also been hearing that major DSPs will implement a remix-like, mash-up-style feature for a while now. What happened to MySpace, however, was different because it pushed a whole new set of artists to the forefront. It gave us the likes of Lily Allen, now more famous for her feet, and Arctic Monkeys. Then we got Soundcloud Rap, based around mixtape culture. Finally, we got TikTok mayhem.
Somewhere along the way, we all tried to make sense of what was happening and how we should listen to these new artists popping up. Anything related to hip hop gets put into the perspective of sampling culture, for example. But sampling and beat-making in the 1980s and 1990s was a vastly different exercise than it was in the 2010s or than it is now. And yet, like with sampling, new creation and distribution channels tend to open up new artistic genres. Perhaps a main difference now with the 1980s is that back then it was about expanding sonic ability, whereas now it’s about avoiding copyright traps.
Content ID
Sampling never went away. Even in a lot of viral memes, sampling plays a major role. And as you well know, the law is always playing catch up with technology. You probably also know what YouTube’s Content ID is, but here’s their own definition:
“Using a database of audio and visual files submitted by copyright owners, Content ID identifies matches of copyright-protected content. When a video is uploaded to YouTube, it's automatically scanned by Content ID.”
There’s a whole set of criteria you have to meet to be able to work with Content ID. For example, you need to be able to show that you have exclusive rights, and - importantly - that you have a large amount of IP that is regularly uploaded to YouTube. Major labels are a good example of meeting these criteria.
A lot of these copyright issues were originally heavily focused on US law, where there exists the notion of Fair Use and which talks about ‘safe harbors.’ At first, platforms like YouTube could take advantage of these safe harbor provisions, because they could claim they, as the platform, were not responsible. There’s a whole rabbit hole here if you want around how this particular American DMCA law has played catch-up if you want. For now, let’s see how this plays out with memes.
Content ID & memes
Memes are particularly difficult to copyright, because they basically exist by virtue of chance and happenstance. Someone makes something, someone else picks it up, the next person broadcasts it to a larger audiences and from there it snowballs. It’s difficult to fix an author to this. Michael Soha and Zachary McDowell showed this in relation to the Harlem Shake meme.
“Just like the original Harlem Shake dance, the Harlem Shake meme was a collectively produced cultural phenomenon. In short, the memetic nature of digital culture is driven by the process of production and sharing, rather than a finished, authored product.”
In other words, there’s no single author and no one potential benefactor of revenue. This is a fully distributed creation, in the case of Harlem Shake this goes back to when the actual Harlem Shake came up in Harlem in the 1970s. There’s no way of telling who started that dance either. What we can say, is that Content ID fails to capture the memetic energy of a craze like the Harlem Shake reborn. Basically, in subverting copyright law, every ‘creator’ that furthers the meme only helps to further concentrate revenues into the hands of the very few who actually run the platforms.
Subverting the beat
As with sampling, so with Content ID. On the one hand, it helps bring cultural relevance to existing, legacy, IP. On the other hand, just like the original drummer of the Amen break never saw a penny of its sampled success, so have very few seen any revenue from Harlem Shake. And just as MySpace, Soundcloud, TikTok, etc. have brought their genres, so has Content ID. It helped spawn the ‘type beat’, a way to move around the Content ID system, yet also take advantage of it.
It helps people find your beat, because you’ve made the beat with the name of a famous producer in the title or deeper down in the metadata. It is, however, an original creation - it’s a beat in the style of. Some even go so far as to find ways to get their ‘type beat’ registered with Content ID to find out that someone else has stolen their beat. See, for example, this exchange on Reddit.
Airbit is one of those beat marketplaces, which evidently fits the YouTube criteria for eligibility for Content ID. The original beat is one by Tyga, which MattFirenzeBeats tried to replicate, or sound like. On sites like Airbit, rappers can license these beats to rappers. Often, these beatmakers start out on YouTube or another social media platform and try to sell beats through there. Like YoungKio, the beatmaker behind Lil Nas X’s Old Town Road. He saved up through direct sales on YouTube:
“Kio would upload his tracks on YouTube with his email in the description for inquiries. Over time, he cobbled together the money for a BeatStars pro account”
That site is where Lil Nas X found the beat and bought it, and YoungKio made the smart move to get a future cut of revenues. This beat may have been original, but it was created in the same system as those type beats of MattFirenzeBeats.
What Content ID has done, then, is to let a whole generation of beatmakers flourish by emulated the styles of others. In doing so, they can subvert the system and stay within its, often fraught, boundaries. All the while, the whole system is unable to account for how what we still call virality actually works. Everything that ‘wins’ on open social platforms, only wins because there’s a whole cabal of people behind it. There is no way to properly assign authorship to everyone involved in generating popularity.
AI Slop sloppiness
It’s pretty clearcut by now, generative AI systems have been trained on copyrighted material. We will see more and more major licensing deals take place. While ‘type beats’ have definitely flooded the likes of YouTube and Soundcloud over the years, it’s nothing compared to what AI Slop is doing right now. If you haven’t witnessed it yourself, it may just be because you’re off social media. 404 Media has been doing some great reporting on AI Slop recently and is a good resource to get stuck in to. A few years ago already, YouTube kids content always had these weird titles (Safari Animals Finger Family Song, anyone?) that played into, or created, weird fads. Now, we see AI Slop playing into these mechanics. From that 404 article, there’s a whole lot, but the strongest one is this: Dora the Explore Feet Mukbang.
Each step seems to take us further away from actual creativeness. Whether it’s a type beat or straight up flooding the zone with AI Slop, who is trying to make something original here? All the while, we’re escaping into our closed spaces online. With good measure, of course, because who wants to have AI Slop jammed down their eyes and into their ears continuously? However, there must be creativity here. There was on MySpace, there was and is on Soundcloud, and there is originality still in type beats.
Perhaps we just still haven’t learned, and we’re thinking about authenticity and originality within the frameworks of copyright as we know them - historically strong, but presently behind the times. Who to credit for AI Slop? Let’s agree that we shouldn’t credit any person for this, but that if the stupid video or image starts to generate revenue we do suddenly want a system in place to bring that revenue to the original idea - Dora the Explorer creators, for example, or whoever.
Retreating into our dark forests we can hopefully see the multitudes at work. This is again about the gray zone. There is a stark need to understand the world better in its complexity. And, to understand people better in their complexity. For a while, it seemed we could do this through open social platforms. Right now, it feels like a flood of AI Slop means we desperately need to find our closed off places. Whether it’s a chat group or a listening bar, this is where we will be able to be original, to be authentic, and to find places to listen to and gaze into our assemblages and try to make some sense out of it all.
LINKS
👍 Was Bandcamp the answer all along? (Declan McGlynn)
“What if—and bear with me on this—the answer isn’t yet another subscription tier, SaaS middleman, or walled garden data gatekeeper promising a superfan utopia? What if you could simply sell the music you make to those who want to buy it?”
✘ It’s not a new argument at all, of course (see this Network Notes from August last year, for example). However, it’s made eloquently again here and I fully support it. Building something new is definitely not always the answer.
🎧 Why listeners interpret songs differently than songwriters intended (Carlo Kiksen)
“The listening environment significantly impacts information processing. Higher cognitive load (like listening while exercising or working) makes deep lyrical processing less likely. In these situations, existing schemas dominate how listeners interpret meaning.”
✘ I always love discussions about the meaning of lyrics. Of course, I rarely have these discussions with the songwriter. That said, the freedom of interpretation, indeed based on schemas of culture and experience like Carlo says, is one of the things that makes music so beautiful to both experience and share.
🪨 Make something heavy (Anu Atluru)
“So many creators start in the shallow pool of some algorithm’s grip—until, inevitably, they go searching for something heavier. From short-form to long-form. From building in public to locking in to solitude, obsession, deep work. To create a book, a film, an album, a company—something that stands alone.”
✘ So, so true. Here I am, putting out this newsletter every week. And yet, it’s all quite light. You read it, perhaps share it, and then you let it go. There’s always a desire to create something heavier. A great piece.
🔐 Open models, closed networks (Tony Lashley)
“Not only does the flooding of traditional networks with AI-generated information cause us to crave the psychological values of community, status and identity more, but these psychological values are becoming more important as AI is making it easier and easier to create software products.”
✘ This piece partially inspired the above piece I wrote. The value of open networks is changing and the value of closed networks is being redefined in that light. How we find people, music, artists, etc. will change with these effects.
🔩 The lo-fi art and human tools era (Pirijan Keth)
“More so now than ever before, making something yourself that you fully understand is deeply empowering. As the world gets noisier, glossier, and more fake, the more valuable the human-touch becomes in both the real world and on the real web. LLMs are the new normal. “Easy to use” became “let me tell you what to write and how to draw”, “here’s what you’ll watch next”, and “here’s what you believe”. But there is always be a market for understandable, engaging, and fun-to-use tools.”
✘ There’s an thread running through the LINKS this week, which focuses on building something where humans thrive, where creativity thrives, and where fun can be had. Perhaps it already exists. Bandcamp exists, and so do analogue cameras, right? It’s all about exploration and focusing your energy away from being told what and how but instead figuring it out yourself and failing along the way.
MUSIC
I need more of this filthy bass in my life. Samurai Breaks just released this record of booty-shaking bass tracks. If you have more of this type of music, please do share it with me. If you want more of this, do follow Opium Hum’s Hyper Real Radio (Telegram or Instagram).