Italy’s Prime Minister Giorgia Meloni has applied for a trademark on her voice, seeking legal protection against its misuse as artificial intelligence makes it increasingly easy to create convincing audio deepfakes. Meloni submitted the application to the European Union Intellectual Property Office (EUIPO) along with an audio recording of about four seconds. In the recording, she says in Italian, “Lo sono Giorgia”, meaning “I am Giorgia Meloni”, twice. Meloni’s move comes amid growing concerns that AI-generated audio can be used to make public figures appear to say things they never actually said. A four-second recording submitted with the application The phrase used in Meloni’s recording is closely associated with her public image. Her 2021 autobiography was titled Io sono Giorgia, and the phrase has also featured prominently in her political campaigning. If approved, the trademark would give Meloni additional legal protection over the commercial use of her voice and related identity. How AI can clone a person’s voice AI tools can analyse recordings of a person’s real voice to learn characteristics such as tone, pronunciation and speaking patterns. The technology can then generate new sentences that sound as if they were spoken by that person. Even a short recording can potentially be used to create a voice imitation, although more high-quality audio generally allows for more convincing results. This has raised concerns about fake statements being circulated in the names of politicians, celebrities and ordinary people. Other celebrities are also seeking protection Meloni is not the only public figure seeking legal protection as AI-generated content becomes more common. American actor Matthew McConaughey applied in January to trademark his image and voice. Singer Taylor Swift also filed three trademark applications in the US in April, covering photographs and voice samples associated with her programmes and previous albums. In August, more than 80 artists urged UK Prime Minister Andy Burnham to introduce legislation giving people legal rights over the use of their voices. Meloni has faced deepfake abuse before The Italian PM has previously been targeted by deepfake content. In 2024, fake pornographic videos using Meloni’s image were circulated online. She later sought €100,000 in damages from those responsible for creating and distributing the videos. In May this year, Meloni also raised concerns about AI-generated images showing her in her underwear that were circulating online. At the time, she said deepfakes were not just a personal problem but a technology that could be used to deceive people, damage reputations and target individuals. Meloni also pointed out that while she had the resources to defend herself, many people did not have the same ability to respond to such misuse. Her latest move to protect her voice reflects a broader shift in how public figures are trying to protect their digital identity in the age of AI.