Italy’s privacy watchdog warns Grok over deepfake AI content

Services that allow users to create and ⁠share content from real images or voices — including digitally "undressing" people — could amount to criminal offences ⁠and serious privacy breaches under EU law, the watchdog ⁠said. It ‌urged providers to build safeguards into their platforms to prevent misuse.

Italy's data protection authority on Thursday warned users ‌and providers of artificial intelligence tools, including Elon Musk’s chatbot Grok, over ⁠the risk of generating deepfake images from real content without the consent of featured individuals. The warning comes amid ​mounting scrutiny of AI platforms for enabling non-consensual ‍sexualized imagery, as EU policymakers push for stricter enforcement under data protection and digital services rules.

The Italian regulator ⁠said ‌it was ⁠working with Ireland’s Data Protection Commission — the lead EU privacy ‍authority for X due to its EU operations being ​based in Ireland — and reserved the right to ⁠take further action. Services that allow users to create and ⁠share content from real images or voices — including digitally "undressing" people — could amount to criminal offences ⁠and serious privacy breaches under EU law, the watchdog ⁠said.

It ‌urged providers to build safeguards into their platforms to prevent misuse.