Spain's People's Party Reportedly Published Purported Deepfakes Featuring Fabricated Admissions of Misconduct

Spain's conservative People's Party reportedly circulated an AI-generated video impersonating the prime minister's wife and Socialist figures making apparent admissions of misconduct. Former Socialist party member Leire Díez filed a defamation complaint; her lawyers alleged insults, anxiety requiring treatment and isolation at home with her daughters. The party reportedly defended the video as political criticism. Its AI disclosure was reportedly small and faint.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.3 Fraud, scams, and targeted manipulation

The incident involved the unauthorized use of public likenesses to impersonate and target specific political figures with defamatory statements for political gain, fitting 4.3.

Additional risk subdomains

• 4.1 Disinformation, surveillance, and influence at scale: The video was a political campaign ad distributed on official party channels to influence public opinion and the upcoming general election.
• 1.2 Exposure to toxic content: The deepfakes falsely depicted individuals admitting to illegal acts like soliciting prostitutes and hiring thugs, exposing the public to toxic/defamatory content and leading to harassment of the victims.

Causal factors

• Entity: Human
• Intent: Intentional
• Timing: Post-deployment

The incident was caused by humans (the People's Party) intentionally using deployed AI deepfake tools to create and distribute a political campaign video targeting rivals.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Risk Level 3: Limited Risk. The report describes 'AI-generated content such as deepfakes' which are subject to transparency obligations. The PP attempted to meet this by noting 'at the bottom – in tiny, nearly transparent font – that it had been created with AI'.

AI system and alleged parties

• AI system: None named
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: Tool
• Alleged developer: Voice cloning technology developers, Synthetic video generation technology developers, Synthetic media technology developers, Synthetic audio generation technology developers, Deepfake technology developers
• Alleged deployer: Synthetic media creators, political organizations, People's Party (Spain), Partido Popular, Deepfake creators
• Alleged harmed parties: Spanish Socialist Workers' Party, Social media users, Santos Cerdán, Public figures, Politicians in Spain, Politicians, political organizations, Partido Socialista Obrero Español, Montse Mínguez, Leire Díez, José Luis Rodríguez Zapatero, José Luis Ábalos, Impersonated public figures, Epistemic integrity, Democratic integrity, Begoña Gómez

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Minor. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Minor, indirect Substantial
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Minor
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Minor
• Epistemic: direct Minor, indirect Minor
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: Yes, the report describes the creation and spread of defamatory deepfake videos.

Directly caused: The PP created and published a 51-second deepfake video falsely depicting political rivals admitting to crimes like soliciting prostitutes, taking kickbacks, and hiring thugs.

Indirectly caused: The video led to a wave of online insults, descalificaciones, and harassment targeting Leire Díez and potentially others.

Democracy

Reported: Yes, the reports describe concerns about the impact of AI-generated disinformation on democratic elections.

Indirectly caused: The use of deceptive deepfakes by a major political party in an official campaign ad erodes democratic norms of truthfulness, fair campaigning, and public trust in political discourse.

Psychological

Reported: Yes, the report explicitly describes psychological distress suffered by Leire Díez.

Indirectly caused: Leire Díez suffered an anxiety attack requiring medical treatment and was forced to isolate at home with her daughters due to public insults and harassment resulting from the video.

Inferred additional harm: The other deepfaked individuals (Gómez, Zapatero, Cerdán) likely experienced significant distress, anxiety, and reputational damage, though specific clinical details are not provided.

Epistemic

Reported: Yes, the reports describe the generation of false information and deepfakes that confuse reality.

Directly caused: The AI system was used to fabricate statements and actions, creating a 'colossal lie' where political figures appeared to admit to crimes they did not commit.

Indirectly caused: The video risks confusing voters who are not familiar with AI, eroding the shared consensus reality necessary for informed democratic voting.

People affected

• Occurrences reported: 1
• People reportedly harmed: 4
• People reportedly exposed: 2000000

Potential causes

Management

• Prioritizing Virality Over Ethics: Party management approved deepfakes to counter rival social media reach.
• Lack of Internal AI Guidelines: No ethical review process within the political party before deploying AI.

Technology

• AI Deepfake Generation: AI was used to manipulate likenesses and voices of political figures.
• Imperceptible Disclaimers: The AI disclaimer was written in tiny, nearly transparent font.

Data Inputs

• Use of Personal Likenesses: Real photos, videos, and voices of politicians were used as inputs.

Human Factors

• Intentional Disinformation: Creators intentionally put false admissions into the mouths of rivals.
• User Confusion: People not used to AI can easily confuse deepfake videos with reality.

Process and Methods

• Inadequate Content Filtering: Social media platforms allowed rapid spread of unverified deepfakes.
• Political Campaign Satire: Using deepfakes as a campaign strategy to dominate online conversation.

Regulatory Environment

• Satire Regulation Loophole: AI regulations contain exceptions for satire, complicating enforcement.
• Lack of Clear Disclosures: Weak enforcement of readable disclosure standards for AI-generated ads.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide consistent, detailed accounts of the incident, including the specific video content, the platform views, the legal complaint filed by Leire Díez, and the political context in Spain. There is no significant ambiguity regarding what occurred or the harms caused.
• Alternative interpretations: The PP argued the video was political satire rather than malicious disinformation, though this does not change the factual occurrence of deepfake generation.

Spain's conservative Peoples Party utilized AI deepfake technology to generate a defamatory campaign video targeting political rivals, including the Prime Minister's wife. While domestic in nature, the incident marks a significant escalation in the use of synthetic media for political disinformation, threatening information security and electoral integrity.

• Overall national security impact: Substantial
• Scope: Single nation
• Primary target: Spain
• Alleged perpetrator: Peoples Party (PP) of Spain

Threat characteristics

• Imminence: Near-term. The active legal dispute and political fallout represent a developing situation requiring monitoring ahead of future elections.
• Autonomy: Human-controlled. The AI system acted purely as a tool under direct human control to generate specific synthetic media.
• Novelty: Evolved capability. While deepfake technology is established, its official deployment by a major political party to defame rivals represents an evolved and concerning campaign tactic.

Impact by dimension

• Physical security: Negligible. No physical threat or critical infrastructure manipulation was involved in this incident.
• Information security: Substantial. Involves sophisticated deepfake operations targeting national leadership and political figures, representing a notable escalation in AI-enhanced political disinformation campaigns.
• Sovereignty: Minor. While targeting high-profile political figures, the incident is a domestic political campaign issue manageable under existing national legal and electoral frameworks.
• Economic security: Negligible. No financial systems or strategic technologies were targeted or compromised in this event.
• Societal stability: Minor. Caused public harassment and psychological distress to an individual, but did not trigger large-scale civil unrest or mass societal instability.