2 million views: Spain's opposition unleashes deepfake ad targeting prime minister's wife

A political advertisement from Spain's conservative People's Party (PP) has generated nearly 2 million views since its release this week. The clip uses artificial intelligence to mimic the voices and faces of figures associated with the ruling Socialist party, including Prime Minister Pedro Sánchez's wife, Begoña Gómez, and former Prime Minister José Luis Rodríguez Zapatero.

In the video, an AI-generated version of Gómez discusses obtaining a university position due to her husband's political role. A deepfake of Zapatero speaks about concealing jewelry in a safe. Both references align with current investigations, though both individuals have publicly denied any wrongdoing. The PP posted the clip to its official social media channels, where it quickly accumulated views and sparked public reaction.

Óscar López, the digital transformation minister, responded to the advertisement on Friday during an interview with broadcaster TVE. He stated that the clip told people a colossal lie and described the use of official party accounts as intolerable. López pointed out that the label indicating the video was AI-generated appeared in tiny, nearly transparent text at the bottom of the screen, making it easy for viewers unfamiliar with the technology to mistake the footage for reality.

Leire Díez, a former member of the Socialist party whose likeness appeared in the ad, filed a court complaint against the PP on Friday. She requested the immediate removal of the video and accused the party of defamation and digital identity theft. The complaint argues that the video used her image to suggest she worked to undermine the judiciary and law enforcement to benefit her former party.

Legal experts consulted by El País generally favored the PP's position. Eight of nine experts agreed the video was not illegal because it was labeled as AI-generated and maintained a satirical tone. Íñigo Navarro, a former dean at Madrid's Comillas Pontifical University, noted that the context made it clear the figures did not actually speak the lines. Juan Antonio Lascuraín, a criminal law professor at the Autonomous University of Madrid, added that legality is distinct from whether the video is considered ethically or politically appropriate.

José Javier Olivas, an associate professor studying populism and AI biases at Spain's National University for Distance Education, called the use of deepfakes by a party claiming to revitalise democracy regrettable. He stated that putting words in other people's mouths qualifies as disinformation and that the video crossed a red line. In the United States, a recent report by a Wesleyan University research group found more than 200 AI-generated ads ahead of next month's midterm elections, with most tied to Republicans. Erika Franklin Fowler, co-director of the Wesleyan Media Project, noted that much of the current AI ad spending carries no disclaimer.

The PP has dismissed the criticism. Party figure Borja Sémper said there would only be a problem if there were malicious intent to make viewers believe the characters were real. He argued that the video clearly uses digital tools to highlight Socialist hypocrisy. Most current polls indicate the PP will win the November 29 election but will require support from the far-right Vox party to form a government.