Skrillex Says AI Songs Can Go Viral But Miss Human Connection

The Human Element Behind the Technology

Before Skrillex became known as a producer and DJ, he was the vocalist of From First To Last, a post-hardcore band that placed him in a scene where live emotion, distortion, and internet-era fandom were closely connected. His move into solo production carried that intensity into songs like Scary Monsters And Nice Sprites and Bangarang, where chopped vocals, aggressive bass design, and software-based production became part of the identity of the record.

In the 032c Issue #49 Summer 2026 interview with Ecco2K, Skrillex discussed making Mora in GarageBand before moving into Pro Tools and Ableton, along with ideas like Bangarang beginning from rough recordings on Snowball mics or laptop mics. Those details show that electronic music does not lose its human side just because it starts from rough digital material, software, or imperfect recordings. The difference is that an artist is choosing what to keep, cut, distort, rebuild, or leave imperfect, while a generated song can imitate the texture of those choices without carrying the bands, scenes, failed ideas, studio accidents, and years of taste that came before the final record.

Why Viral Reach Does Not Equal Real Connection

When Skrillex talks about AI songs going viral, the point is easy to connect to cases the music industry has already dealt with. Heart On My Sleeve, the fake Drake and The Weeknd song made with AI vocals, became a talking point because people were trying to work out how close the vocals sounded, whether the artists had approved it, and what it meant if a fake collaboration could travel that quickly. That kind of attention can make a song look successful on the surface, but the reaction is tied to the stunt, the confusion, and the technology behind it, not to a real connection between artist and listener.

Dance music has its own version of that problem because records often gain meaning before they are officially released. Rumble by Skrillex, Fred again.., and Flowdan was already circulating through DJ set recordings and crowd clips before its official release, while Baby again.. by Fred again.., Skrillex, and Four Tet gained more attention after being played live, including during their back-to-back set at Madison Square Garden. That path is different from an AI song going viral because it sounds close to an existing artist or copies a familiar vocal style. In dance music, the memory around a song is often tied to the DJ who tested it, the crowd that reacted to it, the festival or club clip that kept circulating, and the months of people asking for the release.

What Artists and Platforms Must Clarify

The next stage of AI music will probably come down to what artists, labels, and platforms are willing to make clear before a song reaches listeners. John Summit's call for human-made music labels points to one possible response, especially as fake vocals and copied artist styles become harder to spot at first listen. David Guetta's experiment with an AI-generated vocal in the style of Eminem showed how quickly a technical demo can raise bigger questions about likeness and consent, while deadmau5 has already warned about deepfakes and false endorsement.

AI may keep producing songs that go viral, especially when the hook feels familiar, the voice sounds close to someone famous, or the curiosity around the technology is strong enough to push the clip online. But dance music has always depended on more than a first reaction. People remember the DJ who played a song before release, the set where they first heard it, the crowd clip that made them search for the ID, and the artist history that made the record feel connected to something real. AI can copy parts of that format, but without a person behind the choices, the trust becomes weaker.

The interview appears in 032c Issue #49 Summer 2026.