EU: Agreement to ban AI services that create sexual deepfakes

Member States and the European Parliament reached an statement past nighttime to prohibition successful the EU artificial intelligence services that let radical to beryllium “undressed” without their consent.

On Tuesday, Italian Prime Minister Giorgia Meloni denounced fake images of herself created utilizing artificial intelligence, calling specified deepfakes “a unsafe tool.”

According to the European Parliament, the caller EU prohibition volition use to systems susceptible of generating pornographic images, videos, oregon audio involving children, oregon depicting intimate parts of an identifiable person, oregon showing them engaged successful intersexual activities without consent.

The caller rules volition use from 2 December 2026. From that date, AI systems volition beryllium required to see safeguards to forestall the procreation of specified content.

The measurement was adopted arsenic portion of a revision of the European AI Act, a landmark portion of authorities formally adopted 2 years ago.

On the aforesaid occasion, the 27 EU Member States and MEPs besides agreed—following a connection by the European Commission—to postpone the introduction into unit of caller rules governing alleged high-risk AI systems, meaning those operating successful delicate areas specified arsenic security, health, and cardinal rights.

These rules had primitively been scheduled to instrumentality effect successful August, but the Commission argued that businesses needed much clip to accommodate to this unprecedented regulatory framework.

The main alteration introduced by Member States and MEPs to the Commission’s connection was the constitution of fixed implementation dates, alternatively than the much flexible timetable initially suggested by Brussels.

The caller deadlines are 2 December 2027 for standalone high-risk systems, and 2 August 2028 for those embedded successful different systems oregon products.

The statement comes amid renewed concerns successful the EU implicit the risks associated with artificial intelligence, pursuing caller attraction connected Mythos, a caller exemplary from US startup Anthropic, whose alleged cybersecurity capabilities person raised concerns.