European Parliament agrees to ban AI 'nudifier' apps as deepfakes are deployed to silence women in politics and media

The European Parliament has agreed to ban artificial-intelligence services that "undress" people without consent, RFI reported, after a sharp rise in cases of women in politics, journalism and activism targeted with pornographic deepfakes. From December, AI systems operating in the EU will have to include safeguards preventing the creation of such content, and EU member states have until 14 June 2027 to introduce rules criminalising non-consensual sharing of intimate images, including deepfakes, as well as the creation or manipulation of sexually explicit material without consent.

Nika Kovac, the 33-year-old founder of Slovenia's 8 March Institute and the citizens' initiative My Voice, My Choice, said AI-generated naked images and videos of her began circulating on social media as her campaign pressing the European Parliament and Commission for EU support on abortion access gained traction. Some relatives initially thought one video was real. "I think it was a form of intimidation, meant to make me uncomfortable and stop me continuing to speak about women's rights. This kind of content is another way of silencing women," she told RFI. Kovac described the experience as "psychological torture" and a "very particular way of taking ownership of your body. Placing you in sexual situations without consent, stripping you naked and using your body in this way shows that you are an object, and that you do not matter." She refused to leave social media during the campaign.

French journalist Salomé Saqué said pornographic deepfakes had been deployed as a weapon to "gag, denigrate and humiliate" her, the latest on what she called a "very long list of online violence." Reporters Without Borders has warned of a growing threat to women journalists and cited Argentine journalist Julia Mengolini, founder of radio station Futurock FM, who has condemned a pornographic deepfake falsely portraying her in an incestuous relationship with her brother and filed a complaint against President Javier Milei after he shared a post mocking her efforts to stop the campaign against her. In Italy, the scandal around the Phica website exposed circulation of stolen, altered or sexualised images of well-known women including Prime Minister Giorgia Meloni and opposition leader Elly Schlein; fresh fake images of Meloni in underwear appeared earlier in May. In Germany, the case of actress and television presenter Collien Fernandes -- whose lawyer described it as "the digital Pelicot affair" -- reignited debate over criminalising the creation of such content; she later identified her former husband as the suspected perpetrator. Northern Irish politician Cara Hunter told The Guardian that a pornographic deepfake released before an election nearly ended her career, with her party advising her to stay silent.

Available figures support the pattern. A 2023 study by identity-protection firm Security Hero found 98 percent of online deepfakes were pornographic and 99 percent of those targeted were women. A UN Women report released in late April surveyed more than 600 women in public life and found 6 percent had been victims of deepfakes, 12 percent had had personal images including intimate or sexual content shared without consent, and 41 percent said they self-censored on social media to avoid abuse. The Hubertine Auclert Centre, a French gender-equality organisation, says 45 percent of victims of sexual cyberviolence experience suicidal thoughts or suicide attempts. A 2025 European Parliament report cited a 16-fold increase in pornographic deepfakes shared online over two years.

Inès Girard of the Hubertine Auclert Centre, who helped draft its briefing on the issue, said: "Overwhelmingly, victims are women, including minors." Paris lawyer Rachel-Flore Pardo, who specialises in cyber-harassment and gender-based and sexual violence, said: "The whole dynamic of sexist and sexual cyberviolence is about silencing women and pushing them to exclude themselves from public space and public engagement. The consequences are self-censorship, withdrawal and fear, which leads women to stay silent and retreat." France's 2024 SREN law on securing and regulating the digital space punishes distribution of sexual content generated using someone's image or voice without consent with up to two years in prison and a fine of €60,000, rising to three years and €75,000 when the content is shared online, and opens the way to prosecuting people who share such deepfakes even if they did not create them. Pardo said the law was in place but the question was "how it is applied and what resources are available for investigations," pointing to obstacles in identifying perpetrators, securing platform responses and obtaining full removal of content. The United Nations says fewer than half of countries have laws covering online abuse, with even fewer specifically addressing AI-generated deepfakes.

The European Parliament said pornographic deepfakes had become trivially easy to produce, with dozens of websites and apps able to "nudify" people in a few clicks from a single photo. The Grok scandal -- in which the AI assistant integrated into Elon Musk's X was accused of allowing mass generation of sexualised images of women and minors from real photographs -- triggered a European investigation and underscored how a profile picture is now enough.