Warner Music CEO Robert Kyncl Testifies Before Congress in Support of Federal AI Regulations: ‘The Genie Is Not Yet Out of the Bottle, But It Will Be Soon’

robert kyncl congressional testimony
  • Save

robert kyncl congressional testimony
  • Save
Warner Music Group head Robert Kyncl speaking before the Senate Judiciary Subcommittee on Intellectual Property. Photo Credit: RIAA

Warner Music Group CEO Robert Kyncl has testified before Congress in support of the No Fakes Act (and the No AI Fraud Act), federal legislation designed to target, among other things, unauthorized AI soundalike releases.

Kyncl appeared before the Senate Judiciary Subcommittee on Intellectual Property today, nearly seven months after lawmakers unveiled the bipartisan No Fakes Act. From the outset, we’ve covered the legislation – and the comparatively sweeping No AI Fraud Act, which was introduced at the top of 2024 and per the RIAA “builds on” the older bill.

Of course, these and related proposals – including at the state level, where Tennessee enacted the ELVIS Act last month – have arrived against the backdrop of a quick-evolving AI landscape. And as part of this landscape, auto-generated music (such as unapproved projects made to resemble releases from prominent artists) is making waves. Additionally, industry infringement litigation is ongoing with AI giants like Anthropic.

Enter the No Fakes Act and the No AI Fraud Act, which Kyncl supported in his opening remarks. (Witnesses including FKA Twigs, identified simply as “twigs” by the appropriate name plate, also appeared before Congress. In an opening statement of her own, the artist disclosed that she’s developed a personal AI deepfake.)

In brief, the Warner Music head’s introductory comments didn’t cover any especially groundbreaking details, instead reiterating his backing of the mentioned legislation and specifically urging a framework to license protected media for AI training processes.

Federal legislation, Kyncl proceeded, must enable “each person” to enforce personal name, likeness, and voice rights, besides establishing “meaningful consequences for AI model builders and digital platforms that knowingly facilitate the violation of a person’s property rights.”

Shifting to the congressional hearing itself, lawmakers from both sides of the aisle kicked things off with lengthy remarks about AI soundalike music – including by playing a “Tupac or AI?” audio clip to underscore the highly precise nature of artificial intelligence song outputs and the commercial pitfalls thereof.

Notwithstanding the fact that the majors promptly booted unauthorized AI tracks from dedicated streaming platforms like Spotify – and took aim at a number of non-infringing AI releases – Kyncl expressed the belief that many artists are competing against the machine-generated works for revenue.

“When you have these deepfakes out there, the artists are actually competing with themselves for revenue on streaming platforms. Because there is a fixed amount of revenue within each of the streaming platforms. And if somebody is uploading fake songs of Twigs, and those songs are eating into that revenue pool, there’s less left for her authentic songs.”

Elsewhere in the hearing, Kyncl answered affirmatively when asked whether “new digital replica rights need to be fully transferable,” opining that artists should decide between licensing said rights and transferring them altogether.

Regarding non-negotiable elements of the relevant legislation, Kyncl stressed a need for one’s “consent” when his or her name, likeness, and/or voice are used “to train AI models and create outputs.” And with AI platforms having already ingested all manner of protected media, the exec further emphasized the significance of maintaining detailed training records and the ultimate development of an auto-identification system.

“We are in the unique moment of time where we can still act, and we can get it right before it gets out of hand,” Kyncl relayed of regulating AI. “The genie is not yet out of the bottle, but it will be soon.”