The music industry\u2019s nightmare came true in 2023, and it sounded a lot like Drake. <\/p>\n
\u201cHeart on My Sleeve,\u201d a convincingly fake duet between Drake and The Weeknd<\/a>, racked up millions of streams before anyone could explain who made it or where it came from. The track didn\u2019t just go viral \u2014 it broke the illusion that anyone was in control. <\/p>\n In the scramble to respond, a new category of infrastructure is quietly taking shape that\u2019s built not to stop generative music outright, but to make it traceable. Detection systems are being embedded across the entire music pipeline: in the tools used to train models, the platforms where songs are uploaded, the databases that license rights, and the algorithms that shape discovery. The goal isn\u2019t just to catch synthetic content after the fact. It\u2019s to identify it early, tag it with metadata, and govern how it moves through the system.<\/p>\n \u201cIf you don\u2019t build this stuff into the infrastructure, you\u2019re just going to be chasing your tail,\u201d says Matt Adell, cofounder of Musical AI. \u201cYou can\u2019t keep reacting to every new track or model \u2014 that doesn\u2019t scale. You need infrastructure that works from training through distribution.\u201d<\/p>\n The goal isn\u2019t takedowns, but licensing and control<\/p>\n Startups are now popping up to build detection into licensing workflows. Platforms like YouTube<\/a> and Deezer<\/a> have developed internal systems to flag synthetic audio as it\u2019s uploaded and shape how it surfaces in search and recommendations. Other music companies \u2014 including Audible Magic, Pex, Rightsify, and SoundCloud \u2014 are expanding detection, moderation, and attribution features across everything from training datasets to distribution.<\/p>\n The result is a fragmented but fast-growing ecosystem of companies treating the detection of AI-generated content not as an enforcement tool, but as table-stakes infrastructure for tracking synthetic media.<\/p>\n Rather than detecting AI music after it spreads, some companies are building tools to tag it from the moment it\u2019s made. Vermillio and Musical AI are developing systems to scan finished tracks for synthetic elements and automatically tag them in the metadata.<\/p>\n Vermillio\u2019s TraceID framework goes deeper by breaking songs into stems \u2014 like vocal tone, melodic phrasing, and lyrical patterns \u2014 and flagging the specific AI-generated segments, allowing rights holders to detect mimicry at the stem level, even if a new track only borrows parts of an original.<\/p>\n The company says its focus isn\u2019t takedowns, but proactive licensing and authenticated release. TraceID is positioned as a replacement for systems like YouTube\u2019s Content ID, which often miss subtle or partial imitations. Vermillio estimates that authenticated licensing powered by tools like TraceID could grow from $75 million in 2023 to $10 billion in 2025. In practice, that means a rights holder or platform can run a finished track through TraceID to see if it contains protected elements \u2014 and if it does, have the system flag it for licensing before release.<\/p>\n \u201cWe\u2019re trying to quantify creative influence, not just catch copies.\u201d<\/p>\n Some companies are going even further upstream to the training data itself. By analyzing what goes into a model, their aim is to estimate how much a generated track borrows from specific artists or songs. That kind of attribution could enable more precise licensing, with royalties based on creative influence instead of post-release disputes. The idea echoes old debates about musical influence \u2014 like the \u201cBlurred Lines\u201d lawsuit \u2014 but applies them to algorithmic generation. The difference now is that licensing can happen before release, not through litigation after the fact.<\/p>\n Musical AI is working on a detection system, too. The company describes its system as layered across ingestion, generation, and distribution. Rather than filtering outputs, it tracks provenance from end to end.<\/p>\n \u201cAttribution shouldn\u2019t start when the song is done \u2014 it should start when the model starts learning,\u201d says Sean Power, the company\u2019s cofounder. \u201cWe\u2019re trying to quantify creative influence, not just catch copies.\u201d<\/p>\n Deezer has developed internal tools to flag fully AI-generated tracks at upload and reduce their visibility in both algorithmic and editorial recommendations, especially when the content appears spammy. Chief Innovation Officer Aur\u00e9lien H\u00e9rault says that, as of April, those tools were detecting roughly 20 percent of new uploads each day as fully AI-generated \u2014 more than double what they saw in January. Tracks identified by the system remain accessible on the platform but are not promoted. H\u00e9rault says Deezer plans to begin labeling these tracks for users directly \u201cin a few weeks or a few months.\u201d<\/p>\n \u201cWe\u2019re not against AI at all,\u201d H\u00e9rault says. \u201cBut a lot of this content is being used in bad faith \u2014 not for creation, but to exploit the platform. That\u2019s why we\u2019re paying so much attention.\u201d<\/p>\n Spawning AI\u2019s DNTP (Do Not Train Protocol) is pushing detection even earlier \u2014 at the dataset level. The opt-out protocol lets artists and rights holders label their work as off-limits for model training. While visual artists already have access to similar tools, the audio world is still playing catch-up. So far, there\u2019s little consensus on how to standardize consent, transparency, or licensing at scale. Regulation may eventually force the issue, but for now, the approach remains fragmented. Support from major AI training companies has also been inconsistent, and critics say the protocol won\u2019t gain traction unless it\u2019s governed independently and widely adopted.<\/p>\n \u201cThe opt-out protocol needs to be nonprofit, overseen by a few different actors, to be trusted,\u201d Dryhurst says. \u201cNobody should trust the future of consent to an opaque centralized company that could go out of business \u2014 or much worse.\u201d<\/p>\n