Sony is engaged in an escalating battle against AI-generated deepfake songs that closely mimic some of its most renowned artists, including Harry Styles and Beyoncé. The company has already removed over 75,000 fake tracks from streaming platforms, but it believes this is only a small fraction of the total number of unauthorized AI-generated songs circulating online.
The widespread emergence of deepfake music is posing a serious commercial threat to legitimate recording artists. In an official statement to the UK government, Sony emphasized that these AI-generated songs are causing direct financial harm to artists, particularly those based in the UK. The government is currently reviewing potential copyright law revisions that could permit AI companies to train their models using copyrighted music without prior consent.
While AI-generated images and music have been praised for reducing production costs, concerns are mounting over the quality and authenticity of AI-produced content. Unlike text-based AI models such as ChatGPT—which must maintain factual accuracy—audio and visual AI models can produce more flexible and varied outputs. Supporters of AI-generated music argue that the technology could help democratize content creation, but critics fear a flood of low-quality, algorithmically tailored music that could dilute genuine artistic expression.
A pivotal moment in the debate came in 2023 when an AI-generated track featuring the voices of Drake and The Weeknd went viral. The song sparked alarm across the music industry, as many questioned whether audiences could—or even cared to—distinguish between real and AI-generated music. This raises the troubling possibility that fewer real artists will emerge to create new music, ultimately leaving AI models to train on their own recycled outputs.
UK Prime Minister Sir Keir Starmer has expressed a strong interest in positioning the country as a leader in AI development. His proposal includes allowing AI companies to freely train their models on a wide range of copyrighted content, not just music. Under this plan, companies like Sony would need to opt out actively—a process the company argues would be overly complex and burdensome.
Although some musicians have signed agreements permitting their voices and likenesses to be used in AI-generated works, they remain a minority. In the UK, protests have erupted in opposition to the proposed regulations, with artists arguing that the new framework would make it nearly impossible to enforce copyright protections effectively.
Beyond the music industry, deepfake AI technology is raising broader ethical concerns. One of the most alarming trends is the rise of hyper-realistic nude deepfakes, where individuals’ faces are seamlessly superimposed onto explicit images without their consent. This issue has become particularly prevalent in U.S. high schools. Meanwhile, AI-generated deepfake audio is increasingly being exploited in sophisticated phishing scams, adding another layer of risk to the growing capabilities of artificial intelligence.
As Sony and other industry giants push back against the rise of deepfake music, the battle over AI-generated content is only beginning. The outcome of this fight could shape the future of the music industry—and the role of artificial intelligence in creative expression—for years to come.