A 51-second AI-generated video released by Spain’s opposition Popular Party has prompted a legal complaint and a dispute over the visibility of its disclosure. The PP describes the production as political satire. Critics argue that the small notice does too little to distinguish fabricated speech from a recording of the people depicted.[2][1]
The video circulated during October ahead of the 29 November general election. It portrays figures linked to the Socialist Party, including Begoña Gómez and former prime minister José Luis Rodríguez Zapatero, appearing to speak about political controversies. The dialogue is synthetic; the statements were not made by the depicted people.[2][1]
Is a disclaimer enough when it is nearly invisible?
El País describes the AI notice as small and difficult to see. Its existence is therefore only part of the dispute: the parties also disagree about whether viewers can recognise the production as fabricated political satire.[1]
El País interviewed more than half a dozen legal specialists, who stressed the broad protection given to political satire. Several considered the party authorship, electoral context and visible parody relevant to whether viewers were being deceived. The adequacy of the AI notice was a separate point of disagreement.[1]
Those comments describe the lawyers’ assessments, rather than a judicial determination. They also distinguish several questions: whether a scene is recognisable as parody, whether it falsely attributes specific conduct, and whether it infringes personal-image or reputation rights. A conclusion on one question does not settle the others.[1]
Former Socialist member Leire Díez has filed a legal complaint about her portrayal, according to El País. The filing alleges harm; it is not a finding of liability. The PP maintains that the production is satire.[3][1]
A separate AI political montage attracted nearly three million views on the measure reported by El País earlier in the week. That figure concerns the earlier montage, not a verified audience for this 51-second video. It helps explain the attraction of synthetic imagery to campaign teams seeking rapid attention.[4]
What a viewer can actually identify
The practical disclosure test is straightforward: can someone recognise the scene as fictional in the form in which they actually see it? A label needs to remain legible on a phone, in a cropped repost and during a fast scroll. Identifiable authorship and the surrounding satirical context supply further information.[1]
The lawyers cited by El País drew a sharper distinction for anonymous material that imitates leaked footage. There, the audience may lack both an identifiable author and an obvious signal of parody. That is a different deception risk from a party openly publishing a caricature of its opponents.[1]
The immediate dispute remains unresolved. Its lasting editorial question is specific: does the notice tell an ordinary viewer, at the moment of viewing, that the apparent recording was invented?
Sources & further reading
- El vídeo del PP contra el PSOE agita el debate sobre los límites de la IA en la batalla política El País · 23 Oct 2026 · Accessed 24 Oct 2026
- El PP da un paso más en su campaña y suplanta la identidad de Begoña Gómez, Zapatero y Ábalos El País · 22 Oct 2026 · Accessed 24 Oct 2026
- Leire Díez se querella contra el PP por el vídeo generado con IA para atacar al PSOE El País · 23 Oct 2026 · Accessed 24 Oct 2026
- El beso de Sánchez con Ábalos que nunca existió anticipa la guerra de la IA el 29-N El País · 21 Oct 2026 · Accessed 24 Oct 2026