machine learnings
This story was published as part of Billboard’s music technology newsletter ‘Machine Learnings.’
Sign up for ‘Machine Learnings,’ and Billboard’s other newsletters, here.
Last week, Universal Music Group filed a $500 million lawsuit against TuneCore and its parent company Believe over alleged copyright infringement of UMG’s recordings. The lawsuit presented two core issues: first, that bad actors used TuneCore to upload songs to streaming services that were simply sped up or remixed versions of UMG-copyrighted recordings, often listed under slight misspellings of the real artist, like “Kendrik Laamar” or “Arriana Gramde.” Second, it claimed that “Believe has taken advantage of the content management claiming system” on YouTube “to divert” and “delay… payment of royalties” that belong to record labels.
If you’ve been following the issues in this case over the last few years, this lawsuit feels like a long time coming, and the issues that UMG raises are certainly not just a TuneCore-specific issue — they’re an industry-wide DIY distribution issue. With the vast scale of songs being uploaded through these companies, and staffs that are too small to catch every bad actor, infringing material has, according to just about everybody, flooded onto streaming services.
The distributors know it’s a problem, too. It’s why TuneCore, DistroKid, CD Baby, Symphonic, Downtown and more formed the Music Fights Fraud coalition in 2023 and say they have increasingly invested in preventing fraud and infringement. Unfortunately, Beatdapp, the industry leader in identifying streaming fraud, believes the problem has only worsened since then. UMG is also not convinced that TuneCore is doing enough, saying that the company’s business model incentivizes them to “turn a blind eye” to this damaging activity.
Below, I’ve condensed some of the arguments I’ve heard among industry leaders both for and against DIY distribution continuing just as it is today. I’ll let you judge which outcome is better.
Trending on Billboard
Argument #1: Why its essential to protect DIY distribution as is
It’s easy to take for granted today that anyone who wants to release a song can do it themselves, but that wasn’t always the case. When physical records reigned supreme, record label contracts often favored the companies involved, and seldom went the artists’ way. At the time, artists were essentially forced to sign to a record label if they wanted a chance at shelf space in stores — especially worldwide. This left artists vulnerable to unequal label deals that locked them in for many albums while the label took the lion’s share of the royalties and the copyrights, often in perpetuity.
When Distrokid, CD Baby, TuneCore and the like emerged in the 2000s, they let anyone sign up for distribution services to digital outlets like the iTunes Store for a flat fee and forever altered the power dynamic. Today, the playing field has leveled significantly: hobbyists can get their music out to the world and artists with professional aspirations can wait as long as they want before they have to give up a single percentage point of their master recordings to a label. These companies helped shift negotiating power to the artists, and for the first time, started the process of allowing music fans to decide what songs would pop, rather than the labels that pulled favors with the gatekeepers who worked in radio, retail and the press.
The shift also presented a new, lucrative business opportunity. Music companies no longer need superstars in their catalogs to make their numbers. In fact, they don’t need catalogs at all. A company can now make money by providing services, like distribution, to the masses of previously-overlooked musical hopefuls instead, relying on volume to make up the numbers.
But that volume allowed for the proliferation of fraud, which is a problem that evolves every day, and bad people will always find loopholes. Already, most distributors have implemented common-sense regulations and checks to curb fraud and invested money into quality control teams. But for many experts, it feels impossible to totally solve the problem. As it’s commonly said, this is an endless game of “wack-a-mole.”
But if the barriers to DIY distribution are too significant — like limiting the number of releases, gating who can use it, hiking the platform fee, adding a streaming threshold, or slowing down release time — it could take power away from indie musicians that they have become accustomed to. Such a move would be a step backward for artistic freedom, and the cost of implementing these regulations could threaten to put some of the smaller distributors out of business. Less choice and competition in DIY distribution isn’t better for users.
It’s impossible to put the DIY distribution genie back in the bottle. Artists, who have become used to the current system, would still find ways to get their music out there quickly and cheaply — whether fraudulent or not. Likely, that music would go out on social media or to social-streaming hybrids like YouTube and SoundCloud, both of which pay out royalties and can still be cheated. Streaming services, like Spotify, Apple and Amazon, would risk losing listenership and music discovery to social media platforms — something they already struggle with in today’s TikTok era — and it might not even solve the problems it targeted.
Argument #2: Why the DIY distribution system is in need of serious reform
Currently, over 120,000 songs are uploaded to streaming services every day, a rate that has rapidly increased for years and will likely continue to do so. This is mostly due to DIY distributors. While it is great that aspiring artists can get their music out there cheaply and easily, this has also led to rampant fraud and copyright infringement that puts excessive burdens on rights holders to police their own catalogs online. What happens when we inevitably get to a point where 1 million songs are uploaded every day? We can’t keep going as we are now, and we are in need of serious reform.
While DIY distributors have announced initiatives like Music Fights Fraud and have hosted panels at industry conferences to explain the new methods they are using to stop bad actors, some people say these companies have an incentive for at least some of it to slip by their watch, given their business models rely on receiving fees in exchange for uploading as many songs as possible. Self-policing is not enough, considering this problem only seems to get worse.
The introduction of generative AI has made this matter even more pressing. While it’s impossible to know how much of the music being uploaded today is AI-generated, and to date the streaming services have no regulations against this, it is certainly contributing to the rising number of songs released to streaming services per day. AI songs are believed to be exploited by bad actors to commit streaming fraud, as we saw in the September lawsuit which alleged a musician named Michael “Mike” Smith stole $10 million in streaming royalties by uploading AI-generated songs using a distributor and then used bots to stream them. Bad actors upload AI songs en masse to spread out artificial streams and make their schemes tougher to detect.
It’s hard to argue that it makes a user’s streaming experience better when a platform has a vast number of AI songs and tracks that not a single person has streamed, and it’s clear that these songs, largely stemming from DIY distributors, are diluting the royalty pool at the expense of what some stakeholders have called “professional artists.” The negligibly low payments earned by hobbyists who have accrued hundreds or just a few thousand streams are sometimes lower than the fees one would incur from transferring the royalties into their bank account.
These distributors, the argument goes, should be penalized for the bad actors they let through. This has been proposed in many forms so far, including a financial penalty instituted by streaming services, requirements for significant “know your customer” checks to slow down uploads and verify users’ identities, a minimum stream count threshold before artists can be eligible for royalty collection, a limit to the number of songs a user can upload at a time, an additional fee for storing massive uploads to streaming services, and more.
It’s not a viable business if you rely on a massive scale of song uploads but can’t afford the proper staffers and tools to police them.
In July, popular influencer/podcast host Tinx took to TikTok to ask her followers a question: “Are labels and artists asking random people to make content about music and not say[ing] it’s an ad?” The answer in the over 700 replies to the video was a resounding and simple “yes.”
“Sound campaigns” have been an integral part of music marketing since TikTok took off in 2019, but they differ from other paid promotion campaigns on social media. Captioning a video with #ad, or another similar disclosure, is required by the Federal Trade Commission (FTC) when companies “pay you or give you free or discounted products or services” in exchange for featuring their product in a video, but that has never been the standard for the paid promotion of a song. “Any essence of perceived authenticity can be stripped away when a creator tags a video as paid,” says one digital marketing agency CEO.
As a result, one major label marketer believes “75% of popular songs on TikTok started with a creator marketing campaign,” but says that there’s no way to actually track how many of the songs that go viral on TikTok do so organically or are boosted by thousands to hundreds-of-thousands of dollars’ worth of paid promo.
Trending on Billboard
When asked for clarification about whether or not promoting songs in the background of videos requires disclosure, a representative for the FTC said, “While we can’t comment on any particular example, that practice seems somewhat analogous to a product placement… When there are songs playing in the backgrounds of videos, there are no objective claims made about the songs. The video creator may be communicating implicitly that they like the song, but viewers can judge the song themselves when they listen to it playing in the video. For these reasons, it may not be necessary for a video to disclose that the content creator was compensated for using a particular song in the background in the video. We would evaluate each case individually however.”
While it is not, in most cases, an FTC violation to run undisclosed creator campaigns to promote singles on TikTok, Instagram Reels or YouTube Shorts, it remains a little-understood area of music marketing that many music fans are not aware is happening. “The beauty of Tiktok, for me, has disappeared because I’m super cynical and believe everything I see there, disclosed or not, is paid to be promoted,” says the digital marketing agency CEO. (Most of the sources in this story requested anonymity in order to speak freely about how these campaigns work.)
Often, digital marketing gurus will reminisce about the days of the Hype House bros and the D’Amelio’s TikTok reign, around the onset of the COVID-19 pandemic, which were considered the good ol’ days for creator marketing. At the time, it was expected for successful TikTok virality to translate into boosts in streams practically every time. “Back then, it made sense to pay over $10,000 a video for those famous kids to post your song. There was a high probability of [return on investment] ROI in 2020,” says a second digital marketing agency CEO. One creator manager says they remember a top creator at the time boasting about getting “$50,000 to just play the sound” in the background of a TikTok.
Typically, these creators would be instructed by an artist manager, a label, or a third-party digital marketing company (most times the latter) to perform a certain trend along with the song, like a dance or a certain filter, in exchange for money.
But these days, experts like George Karalexis, CEO of YouTube marketing and rights management company Ten2 Media, say it’s “more expensive and harder than ever to start a trend” online. As Billboard reported in 2022, TikTok tracks in the U.S. were streamed far less that year than they were in 2021, according to the most recent available data from Luminate.
Now, this unpredictability has led to top creators rarely fetching rates of over $10,000 for the use of a song in a video. Instead, digital marketers are spreading their budgets over many videos from smaller creators to make the illusion of a less-detectable groundswell of support. The second digital marketing agency CEO says today’s payment ranges from $25 for a micro creator (at or below about 10,000 followers) to $10,000 for a TikTok star to post the song.
Recently, a cottage industry of startups has popped up in the creator campaign space, automating the connection between smaller creators and artists looking to pay them to promote their songs. One of the leading companies, Sound.Me, for example, recently ran a creator campaign for “A Bar Song (Tipsy)” through their service. TikTok is also offering up a similar service with its “Work With Artists” feature inside the app, which allows qualifying creators (those with over 50,000 followers and living within a certain territory) to get paid to use songs, like Halsey’s cover of “Lucky,” in their videos.
Even when an artist is willing to spend a significant budget on one particular creator, that doesn’t mean the creator will always accept. Sound promos are known to be less lucrative for creators than other brand deals, like fashion or skincare, and thus it’s common for top creators to “shoot [the artist’s team] an outrageous number, knowing a sound campaign is not necessarily worth their time otherwise,” says the creator manager.
It is also far less common to ask for a specific type of video from a creator today. Instead, the second digital marketing CEO says “it’s not really about pushing specific creative. It’s just about finding the right creators for the artist’s target audience and kind of just letting the influencers run with creative freedom.”
All of this makes discerning the paid promotion of a song from organic enthusiasm more challenging than ever. Even more complicated, the creator manager says that it’s “best practice” for creators “who want to work with a specific brand to show for free that they are using the brand’s products anyway to attract their attention. Same goes for songs.”
The sign of true success for these campaigns is when social media use of the song grows far beyond the initial budget, encouraging unpaid creators to jump in and use the track, too, multiple digital marketing sources say. “The value is in the people [using the song] that aren’t being paid,” says Jeremy Gruber, head of artist marketing and digital strategy at management company Friends at Work. “Success is when we have 13 types of videos going on at once to the song,” adds one indie label marketer. “We can’t even tell what’s happening.”
Typically, these sound campaigns are conducted in phases, and while they are common, they are not expected for every single release, three label marketing sources say. $5,000 is the low-end for what two digital marketing agencies believe would be a fruitful campaign, but the spending can grow to $80,000 (or even into the six figures for rare cases) if it is a big-name artist and the song is reacting positively. Typically after the first round of the campaign, the team will watch and see if the song grows. If it does, then a next wave of spending will be opened up and seeded out to creators to stoke the flame.
Gruber believes an ethical gray area arises when artists’ teams offer money to music curation influencers to explicitly recommend a song without disclosing the transaction to viewers. Unlike a “product placement”-like promotion which simply streams in the background, these music curators use TikTok to talk to the camera, telling consumers to take action and check out new songs in exchange for undisclosed money, concert tickets or other perks. When asked about this type of promotion specifically, the FTC declined to comment on whether or not disclosure is needed.
It’s also common for record labels to turn to social media-based blogs, typically in the rap genre, like WorldStarHipHop, Rap, Our Generation Music and more which offer pay-to-play promotion on TikTok and other social platforms to create the appearance of organic online chatter. In one message exchange, reviewed by Billboard, a representative from Rap told a music company that “solo” posts go for $1,000, but they offer discounted rates for ordering in “bulk.” Typically, these payments are not disclosed to consumers.
While it might come as a surprise to some music lovers to learn how often these paid campaigns are used, the general consensus among the eight sources spoken to for this story is that it isn’t harming anyone to do it —at least not in the types of campaigns that resemble product placements. “Music, to me, is this beautiful art form and it is completely different from other ‘products’ in other industries [that run creator campaigns],” says the first digital marketing agency CEO. “We do feel that ethically we’re promoting content that is a net positive to society.”
It may not be as effective as it was a few years ago, but creator campaigns are largely believed to still be essential to market songs today, whether it’s on TikTok or on Instagram Reels or YouTube Shorts (which is increasingly common). Says the second digital marketing agency founder: “It’s still the best thing we have.”
This story was published as part of Billboard’s new music technology newsletter ‘Machine Learnings.’ Sign up for ‘Machine Learnings,’ and Billboard’s other newsletters, here.
On Sept. 4, the public learned of the first-ever U.S. criminal case addressing streaming fraud. In the indictment, federal prosecutors claim that a North Carolina-based musician named Michael “Mike” Smith stole $10 million dollars from streaming services by using bots to artificially inflate the streaming numbers for hundreds of thousands of mostly AI-generated songs. A day later, Billboard reported a link between Smith and the popular generative AI music company Boomy; Boomy’s CEO Alex Mitchell and Smith were listed on hundreds of tracks as co-writers.
(The AI company and its CEO that supplied songs to Smith were not charged with any crime and were left unnamed in the indictment. Mitchell replied to Billboard’s request for comment, saying, “We were shocked by the details in the recently filed indictment of Michael Smith, which we are reviewing. Michael Smith consistently represented himself as legitimate.”)
Trending on Billboard
This case marks the end of generative AI music’s honeymoon phase (or “hype” phase) with the music industry establishment. Though there have always been naysayers about AI in the music business, the industry’s top leaders have been largely optimistic about it, provided AI tools were used ethically and responsibly. “If we strike the right balance, I believe AI will amplify human imagination and enrich musical creativity in extraordinary new ways,” said Lucian Grainge, Universal Music Group’s chairman/CEO, in a statement about UMG’s partnership with YouTube for its AI Music Incubator. “You have to embrace technology [like AI], because it’s not like you can put technology in a bottle,” WMG CEO Robert Kyncl said during an onstage interview at the Code Conference last September.
Each major music label group has established its own partnerships to get in on the AI gold rush since late 2022. UMG coupled with YouTube for an AI incubator program and SoundLabs for “responsible” AI plug-ins. Sony Music started collaborating with Vermillio for an AI remix project around David Gilmour and The Orb’s latest album. Warner Music Group’s ADA struck a deal with Boomy, which was previously distributing its tracks with Downtown, and invested in dynamic music company Lifescore.
Artists and producers jumped in, too — from Lauv’s collaboration with Hooky to create an AI-assisted Korean-language translation of his newest single to 3LAU’s investment in Suno. Songwriters reportedly used AI voices on pitch records. Artists like Drake and Timbaland used unauthorized AI voices to resurrect stars like Tupac Shakur and Notorious B.I.G. in songs they posted to social media. Metro Boomin sampled an AI song from Udio to create his viral “BBL Drizzy” remix. (Drake later sampled “BBL Drizzy” himself in his feature on the song “U My Everything” by Sexyy Red.) The estate of “La Vie En Rose” singer Edith Piaf, in partnership with WMG, developed an animated documentary of her life, using AI voices and images. The list goes on.
While these industry leaders haven’t spoken publicly about the overall state of AI music in a few months, I can’t imagine their tone is now as sunny as it once was, given the events of the summer. It all started with Sony Music releasing a statement that warned over 700 AI companies to not scrape the label group’s copyrighted data in May. Then Billboard broke the news that the majors were filing a sweeping copyright infringement lawsuit against Suno and Udio in June. In July, WMG issued a similar warning to AI companies as Sony had. In August, Billboard reported that AI music adoption has been much slower than was anticipated, the NO FAKES Act was introduced to the Senate, and Donald Trump deepfaked a false Taylor Swift endorsement of his presidential run on Truth Social — an event that Swift herself referenced as a driving factor in her social media post endorsing Kamala Harris for president.
And finally, the AI music streaming fraud case dropped. It proved what many had feared: AI music flooding onto streaming services is diverting significant sums of royalties away from human artists, while also making streaming fraud harder to detect. I imagine Grainge is particularly interested in this case, given that its findings support his recent crusade to change the way streaming services pay out royalties to benefit “professional artists” over hobbyists, white noise makers and AI content generators.
When I posted my follow up reporting on LinkedIn, Declan McGlynn, director of communications for Voice-Swap, an “ethical” AI voice company, summed up people’s feelings well in his comment: “Can yall stop stealing shit for like, five seconds[?] Makes it so much harder for the rest of us.”
One AI music executive told me that the majors have said that they would use a “carrot and stick” approach to this growing field, providing opportunities to the good guys and meting out punishment for the bad guys. Some of those carrots were handed out early while the hype was still fresh around AI because music companies wanted to appear innovative — and because they were desperate to prove to shareholders and artists that they learned from the mistakes of Napster, iTunes, early YouTube and TikTok. Now that they’ve made their point and the initial shock of these models has worn off, the majors have started using those sticks.
This summer, then, has represented a serious vibe shift, to borrow New York magazine’s memeable term. All this recent bad press for generative AI music, including the reports about slow adoption, seems destined to result in far fewer new partnerships announced between generative AI music companies and the music business establishment, at least for the time being. Investment could be harder to come by, too. Some players who benefitted from early hype but never amassed an audience or formed a strong business will start to fall.
This doesn’t mean that generative AI music-related companies won’t find their place in the industry eventually — some certainly will. This is just a common phase in the life cycle of new tech. Investors will probably increasingly turn their attention to other AI music companies, those largely not of the generative variety, that promise to solve the problems created by generative AI music. Metadata management and attribution, fingerprinting, AI music detection, music discovery — it’s a lot less sexy than a consumer-facing product making songs at the click of a button, but it’s a lot safer, and is solving real problems in practical ways.
There’s still time to continue to set the guardrails for generative AI music before it is adopted en masse. The music business has already started working toward protecting artists’ names, images, likenesses and voices and has fought back against unauthorized AI training on their copyrights. Now it’s time for the streaming services to join in and finally set some rules for how AI generated music is treated on its platforms.
This story was published as part of Billboard’s new music technology newsletter ‘Machine Learnings.’ Sign up for ‘Machine Learnings,’ and Billboard’s other newsletters, here.
If you have any tips about the AI music streaming fraud case, Billboard is continuing to report on it. Please reach out to krobinson@billboard.com.
Even before President Joe Biden announced that he was dropping out of the 2024 presidential race on July 21, extremely online millennials and Gen Zers had started posting memes on social media in support of Vice President Kamala Harris, who many hoped (and assumed) would take over for Biden after his disastrous debate performance in late June. And after Harris replaced him as the presumptive Democratic presidential nominee, it seemed the entire internet became completely coconut-pilled.
Along with traditional text- and image-based memes — which are nothing new — musical memes have also proliferated on short-form video sites like TikTok, Reels and Shorts, with users mashing up Harris quotes with popular songs using AI or more traditional methods of remixing. But these playful — or, in some cases, just plain strange — songs are more than just digital fun and games. The overwhelmingly pro-Harris memes are reaching millions of potential voters, and might help Harris mobilize the previously discouraged young voters she needs in order to win in November.
One audio, which has over 1.1 million likes on TikTok, pairs Harris’ memeable quote “do you think you just fell out of a coconut tree?” with the instrumental for “360” by Charli XCX. Another pitch-alters the same Harris quote over “The Star-Spangled Banner.” One anti-J.D. Vance audio pastes the Republican VP candidate saying “I’m a Never Trump Guy” over “Freek-a-Leek” by Petey Pablo. (After that clip went viral, the @KamalaHQ account also made its own video using the sound.)
There are also pro-Harris AI tracks, like one that replaces the lyrics to a Beyoncé song to make Queen Bey seemingly sing “you exist in the context of all in which you live,” another heavily memed Harris quote. A different AI track splices a Harris soundbite over DJ Johnrey’s viral track “Emergency Budots,” with an AI deepfake video of Harris and Pete Buttigieg dancing under a palm tree.
Beyond its political ramifications, this content also offers a glimpse into the future of music — one where we don’t just play our music, but where we play with it. In a sense, it’s the culmination of a trend that’s been brewing for decades. As music lovers have embraced sampling, remixing, the digital audio workstation, the Splice royalty-free sample library, Kanye West’s stem player and sped-up/slowed-down song edits, they’ve demonstrated a desire to have more control over static recordings than the traditional music consumption provides. And AI innovations can help to further facilitate this customizable listening experience.
Some music AI experts, including Suno’s CEO Mikey Shulman, are betting on a future where “anyone can make music” at the click of a button — and that everyone will want to. Often, I’ve heard folks who espouse this view of AI music compare it to photography, given photography is an art form which went from being something conducted by trained professionals in proper studio settings to being a ubiquitous activity aided by smartphones.
These entrepreneurs aren’t totally misguided — it’s clear based on user interest in Suno and Udio that there is a place for songs that are completely new and individual. But right now, it seems predictions about this technology’s role in the future of music consumption are too bullish. Music fans still crave familiarity, community and repetition when listening to music. It’s also scientifically proven that it takes multiple listens to form bonds with new songs — which is way more likely to happen with hit songs by artists you know and love, rather than individualized AI-generated tracks.
Instead, I think the average music listener will be way more interested in using AI to tweak their favorite hits. Listeners could use AI stem separation tools to create more bass-heavy mixes, for example, or some form of AI “timbre transfer” to make a song’s guitars sound more like a Les Paul than a Stratocaster (you could also go even further and change a guitar to be an entirely different instrument), or AI voice filters to change the lyrics of a song to include their best friend’s name.
Of course, there are still serious legal hurdles to customizing copyrighted sound recordings and songs if users share them publicly. Right now, any of the artists whose songs were used in these pro-Harris remixes could get them taken down upon request, citing copyright infringement. The NMPA has also expressed that it is willing to fight back against Spotify if it ever rolled out customizable song features on its platform. In a cease and desist letter, the NMPA warned the streaming service, saying, “We understand that Spotify wishes to offer a ‘remix’ feature…to ‘speed up, mash up, and otherwise edit’ their favorite songs to create derivative works. Spotify is on notice that release of any such feature without the proper licenses in place from our members may constitute additional direct infringement.”
So for now, edited songs will remain on social media platforms only, at least until they receive takedown requests. Still, consumer interest in music customization is only growing, and the popularity of pro-Harris campaign remixes serve as proof.
This analysis was published as part of Billboard’s new music technology newsletter ‘Machine Learnings.’ Sign up for ‘Machine Learnings,’ and Billboard’s other newsletters, here.
Thirteen years ago, the then-unknown teenager Rebecca Black posted her song “Friday” to YouTube, hoping to spark her music career. We all remember what happened next. The song, which amassed 171M views and 881K comments on YouTube to date, was pushed up the Billboard charts, peaking at No. 58 on the Hot 100. “Friday” was a true cultural phenomenon — but only because it was a laughingstock.
“I became unbelievably depressed,” Black said of the song’s meme-ification — and the cyberbullying that came with it — on Good Morning America in 2022. “And [I felt] trapped in this body of what the world would see me as forever. I hadn’t even finished growing.”
Many music makers dream of waking up one morning and realizing a song of theirs has gone viral overnight. But, as Black’s experience shows, not all virality is created equal. At best, it can bring a Hot 100 hit, radio play and a slew of new, lifelong fans. At worst, it can be the artists’ worst nightmare.
Trending on Billboard
One such worst-case scenario recently took place with Gigi D’Agostino’s 1999 Italo dance track “L’amour Toujours,” which was recently co-opted by the German far-right. In a popular video posted to social media, a group of young men sang the song outside a bar on the German island Sylt, replacing the original lyrics with a Neo-Nazi slogan that translates roughly to “Germany to the Germans, out with the foreigners.” As they chanted the xenophobic lyric, one of the men raised his arm in a Nazi-like salute. Another put two fingers to his upper lip in a seeming allusion to Adolf Hitler’s characteristic mustache.
After that, several events in Germany, including Oktoberfest in Munich, looked into banning the song, and D’Agostino replied to an inquiry from German newspaper Der Spiegel with a written statement, claiming that he had no idea what had happened.
Granted, the circumstances of virality are rarely that bad, but songs commonly end up on an “unintended side of TikTok,” as Sam Saideman, CEO/co-founder of management and digital market firm Innovo, puts it. “We try to educate our partners that sometimes you cannot control what uses of your song [are] on the internet.” While Innovo “may plan a campaign to [pay creators to] use the song in get-ready-with-me makeup videos,” he explains, another user’s totally different kind of video using the song could become far more popular than the originally planned use, pushing the campaign organically onto another part of the platform and away from its target audience.
For example, Twitter and TikTok users twisted “Cellophane,” FKA Twigs’ heartbreaking 2019 ballad about unrequited love, into a meme beginning in early 2022. Oftentimes, videos using the song pair Twigs’ voice with creators that are acting melodramatic about things that are clearly no big deal. Even worse, one popular version of the audio replaces Twigs’ voice with Miss Piggy’s (yes, the Muppet character).
“Digital marketers are able to boost certain narratives they support,” says Connor Lawrence, chief marketing officer of Indify, an angel investing platform that helps indie artists navigate virality. “It happens a lot — marketers boosting a narrative that is most favorable to the artist’s vision to hopefully steer it.” Saideman says he likes to keep a “reactionary budget” on hand during his song campaigns in case they need to try to course-correct a song that is headed in the wrong direction.
But digital marketing teams can’t do much to fix another bad type of song virality: when songs blow up before the artist is ready. “I am actively hoping that my baby artist does not go viral right now,” says one manager who wished to remain anonymous to protect their client’s identity. “They need to find their sound first.” Omid Noori, president/co-founder of management company and digital marketing agency ATG Group, adds, “It’s a real challenge when someone goes viral for something when they aren’t ready to capitalize on it, or even worse, the song that took off sounds nothing like anything you want to make again.”
Ella Jane, an indie-pop artist who went viral in 2020 for making a video that explained the lyrics to her song “Nothing Else I Could Do,” says that going viral early in her artistic career had positive and negative effects. She signed a deal with Fader Label and boosted her following, but she’s also still dealing with the downsides four years later. “I’m grateful for it, but I think because my first taste of having a successful song was inextricable from TikTok, it has cast a shadow on my trajectory in some ways,” she says.
Over her next releases, Jane says she chased the algorithm, like many of her peers who experienced TikTok hits early in their careers, trying out lots of different video gimmicks to hook listeners. “It doesn’t reflect who I am as an artist now,” she says. “That feeling is addicting, and you feel like you’re withdrawing from it when your videos don’t hit. It can leave artists at a point where they’re obsessed with metrics.” This obsession has been reinforced by some record labels who use metrics as the only deciding factor in whether or not to sign a new artist.
“This is no different than hitting the lottery,” Noori says. “Imagine you get the $100 million jackpot on your first try… It makes artists feel like failures before they even really get started.”
As artists are increasingly instructed by well-meaning members of their team to make as many TikToks as possible, some have turned to sharing teasers of unfinished songs as a form of content — which have occasionally gone viral unintentionally, despite not even being fully written and recorded. That’s what happened to songs by Good Neighbours, Leith Ross, Katie Gregson MacLeod and Lizzy McAlpine, leading many of them to rush to finish recordings so they could capitalize on their spotlight before it faded.
“People put a lot of pressure on the recorded version,” says Gregson MacLeod, whose acoustic piano version of her song “Complex” went viral before she had recorded the official master. “If it is not exactly like the sound that went viral, if you don’t sing the words in the exact same way or use the exact same key, sometimes people decide, ‘We’re not having it.’” While she says she was ultimately happy with how it all turned out, not everyone is so lucky. Within two weeks of the song’s virality, she rushed to release a “demo” version to match the rawness of her original video, as well as a produced version, earning her a combined 43 million plays on Spotify alone.
McAlpine, however, decided to run away from her unfinished viral song. After posting a popular video of herself playing a half-written song, she told her fans in a TikTok video, “I’m not releasing that song ever because I don’t like it. It doesn’t feel genuine. It never felt genuine. I wrote it for fun. It wasn’t something I was ever going to release, or even going to finish… That is not who I am as an artist; in fact, I think I’m the opposite… I’m not concerned with overnight success. I’m not chasing that… I want to build a long-lasting career.”
Noori says TikTok virality in particular has led to a “huge graveyard of one-hit wonders,” something that is far more common today than the bygone days of traditional, human gatekeepers. “With the algorithm, how do you even know who saw your content?” he asks.
Still, there’s an argument to be made that perhaps, as P.T. Barnum famously said, “There’s no such thing as bad publicity.” “I’ve been thinking about that idea a lot and whether or not it is true for virality,” says Saideman. “And it’s hard to say.”
Black ultimately reclaimed “Friday” and her music career in 2021 by getting in on the joke, turning the decade-old cult hit into a hyperpop remix, produced by Dylan Brady of 100 Gecs and featuring Big Freedia, Dorian Electra and 3Oh!3. From there, Black continued to release music as a queer avant pop artist and played an acclaimed DJ set at Coachella in 2023. Still, the original version of “Friday” is her most popular song on Spotify by a long shot, even though it was released before the streaming era began.
“The beauty and curse of these platforms, especially TikTok today,” Saideman says, “is that they are remix platforms. When you put your music on them, you are opening your music up creatively to other people using it in positive and negative ways. You can’t have one without the other.”
This story was featured in Billboard’s new music technology newsletter ‘Machine Learnings.’ Sign up to receive Machine Learnings, and Billboard’s other newsletters, for free here.
-
Pages