Deepfake Attack on Prime Minister's Wife Sparks Electoral Firestorm in Spain

Voters in Spain face a new layer of complexity in the coming election cycle, as a political advertisement that uses artificial intelligence to mimic the prime minister's wife has triggered a national dispute. The clip, released by the main opposition party, forces ordinary citizens to navigate the growing gap between digital media and physical reality. This development marks a significant shift in how campaigns communicate, placing the credibility of visual evidence under intense scrutiny just weeks before the general election.

The conservative People's Party distributed the video across its social media platforms this week. It leveraged AI tools to manipulate the images of individuals affiliated with the governing Socialist party. The content appears to show these figures admitting to various scandals, including hiring prostitutes and receiving illicit payments. One specific segment features an AI-created version of Begoña Gómez, the prime minister's wife, discussing how she secured a university position due to her husband's status. Another segment includes a deepfake of former prime minister José Luis Rodríguez Zapatero claiming he hid jewelry in a safe. Both scenarios reference active investigations, though both individuals have denied any wrongdoing.

By Friday, the clip had accumulated nearly two million views. The Socialist party reacted swiftly to the backlash. Óscar López, the minister for digital transformation, told broadcaster TVE that the spot was telling people a colossal lie. He argued that it seemed intolerable for a party to publish such content on its official channels. López pointed out that the label indicating the video was AI-generated was displayed in tiny, nearly transparent text at the bottom of the screen. He warned that viewers unfamiliar with these technologies could easily mistake the footage for genuine recording.

Leire Díez, a former member of the Socialist party whose image was included, filed a court complaint against the People's Party on Friday. She requested the immediate removal of the advertisement. Díez accused the opposition of defamation and digital identity theft. The video used her likeness to suggest she worked to weaken the judiciary and law enforcement to advantage her former party.

Legal analysis generally supports the opposition's position. Most of the nine experts consulted by the newspaper El País agreed the video was not illegal. They cited the clear AI label and the satirical nature of the content. Íñigo Navarro, a former dean at Madrid's Comillas Pontifical University, stated that in this context, everyone understands the people did not actually say those things. Juan Antonio Lascuraín, a criminal law professor, noted that legality is distinct from questions of appropriateness or ethical correctness.

The situation is not unique to Spain. In the United States, a similar trend is emerging ahead of next month's midterms. A report from Wesleyan University identified over 200 ads using AI, most linked to Republicans. Erika Franklin Fowler, co-director of the project, noted that spending on such ads in the previous cycle was under $50,000. She observed that much of the current usage lacks any disclaimer.

The People's Party has dismissed the criticism as excessive. Borja Sémper, a party figure, argued that a problem would only exist if there was malicious intent to deceive. He maintained that viewers can see the digital tools are used to highlight Socialist hypocrisy. Current polls suggest the People's Party will win the November 29 election, though they will likely need support from the far-right Vox party to form a government.