Sony Music has taken decisive action against the growing threat of AI-generated deepfake recordings, removing over 75,000 fake songs from various streaming platforms. Whilst the company has not publicly disclosed specific titles of the removed songs, it has reported issuing close to 10,000 takedown notices for unauthorised deepfakes impersonating its artists. The 75,000 figure refers to the total number of removed items, whilst the 10,000 takedown notices represent legal actions taken against specific cases of deepfake misuse. This move comes as the music industry grapples with the rapid rise of artificial intelligence, raising significant concerns over copyright infringement and artist protection.
Sony Music UK supports the ‘Make It FAIr’ campaign, calling on policymakers to respect UK copyright law and not let Gen AI companies train on music without permission. pic.twitter.com/7CS9iuIM80
— Sony Music UK (@SonyMusicUK) February 25, 2025
In a recent submission to a UK government consultation on copyright laws, Sony Music expressed concerns that potential changes in legislation could exacerbate the problem. Rob Stringer, Chairman of Sony Music Group, said: “AI is going to change the world, however, there can be no question that creators must be rewarded for being part of this technological revolution. Safeguarding the existing UK copyright model ensures AI developers and content owners can innovate whilst also recognising the value of the works on which their products are based”. Several celebrities have publicly addressed the issue of the proposed copyright law changes, including Elton John, Kate Bush, Paul McCartney, and Jimmy Page. These musicians have actively voiced their concerns regarding the unauthorised use of artists’ works in AI training, advocating for stronger protections within the creative industry.
Executives at Sony believe that the 75,000 identified deepfake recordings are only a fraction of the total AI-generated content circulating online. Unlike traditional copyright enforcement, identifying and removing AI-generated material is a complex and time-consuming process that relies heavily on manual detection methods. Artificial intelligence has become an increasingly powerful tool in music production, enabling users to create hyper-realistic vocal imitations of popular artists. From AI-generated covers to entirely new songs mimicking famous voices, the technology has raised ethical and legal questions about intellectual property in the digital age. For example, here is an AI-generated Amy Winehouse singing Miley Cyrus’ ‘Flowers‘ in a jazz style:
Sony’s actions reflect broader concerns within the music industry about the implications of AI-generated content. The company has urged policymakers to implement stricter copyright protections, ensuring that artists maintain control over their work and are fairly compensated for their contributions. Several other major record labels have also voiced their concerns, including Universal Music Group (UMG) which is advocating for AI protections in licensing agreements with distributors like TikTok and Meta. The industry is now calling for more advanced content identification tools and clear legal frameworks to protect artists from exploitation.
As the debate over AI deepfakes intensifies, Sony Music’s efforts highlight the urgent need for regulatory action. With technology continuing to evolve rapidly, the music industry faces an ongoing battle to safeguard artistic integrity in an era of artificial intelligence.
