Purported AI-Manipulated Videos Depict Mauricio Macri Endorsing Manuel Adorni and Silvia Lospennato Withdrawing Before Buenos Aires Vote

Purported AI-generated deepfake videos were reportedly circulated on X in the final hours before the Buenos Aires municipal election in May 2025. The alleged synthetic content depicted Mauricio Macri endorsing Manuel Adorni and Silvia Lospennato withdrawing. Officials reportedly described the material as an attempt to mislead voters. A formal complaint was filed with the electoral tribunal, and an investigation was requested.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.1 Disinformation, surveillance, and influence at scale

The incident involved a coordinated campaign using AI-generated deepfakes of political figures to influence public opinion and manipulate the Buenos Aires electoral process.

Additional risk subdomains

• 3.1 False or misleading information: The deepfakes and fabricated news articles generated false information about candidates withdrawing from the election.

Causal factors

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

The incident was caused by human actors intentionally creating and distributing AI-generated deepfakes to manipulate voters before an election.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The AI systems used were for generating AI-generated content such as deepfakes, which under the EU AI Act are subject to transparency obligations to inform users that the content is artificially generated.

AI system and alleged parties

• AI system: unspecified
• AI purpose: Deepfake Video Generation; Deepfake Image Generation
• Behaviour type: Tool
• Alleged developer: Unknown voice cloning technology developers, Unknown deepfake technology developers
• Alleged deployer: Unknown actors affiliated with pro-La Libertad Avanza social media accounts, @PatricioVogel1, @MileiEmperador, @GordoDan_
• Alleged harmed parties: Voters of Buenos Aires, Voters of Argentina, Silvia Lospennato, Mauricio Macri, General public of Argentina, Electoral integrity

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 Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Minor, indirect Negligible
• Democracy: direct Minor, indirect Minor
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Minor, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: Yes

Directly caused: The incident directly involved the creation and viral spread of AI-generated deepfake videos of Mauricio Macri and candidate Lospennato on X.

Inferred additional harm: It is likely that other similar deepfakes or manipulated media were generated and shared during the campaign, though not explicitly detailed.

Civil rights

Reported: Yes

Directly caused: The deepfakes directly targeted the citizens' right to cast a conscious vote by spreading false information about candidate withdrawals.

Democracy

Reported: Yes

Directly caused: The deepfakes directly interfered with the democratic election process in Buenos Aires by attempting to mislead voters hours before the election.

Indirectly caused: The incident led to formal complaints to the electoral tribunal and requests for judicial intervention, undermining trust in the electoral process.

Epistemic

Reported: Yes

Directly caused: The creation of highly realistic deepfakes and fabricated LA NACION articles misled social media users regarding the status of candidates.

Inferred additional harm: The widespread circulation of such deepfakes erodes public trust in digital media and shared reality during critical democratic events.

People affected

• Occurrences reported: 1
• People reportedly harmed: 3
• People reportedly exposed: 10000

Potential causes

Management

• Unethical Campaign Strategies: Campaign managers leveraged deceptive AI tools to gain electoral advantages.
• Inadequate Platform Risk Control: Platform management failed to prepare for AI-driven election interference.

Technology

• AI Deepfake Generation: AI tools allowed credible recreation of politicians' voices and faces.
• Rapid Social Media Viralization: Platform algorithms enabled the swift spread of the deceptive videos.

Data Inputs

• Public Training Data Exploitation: Existing public recordings of Macri and Lospennato used to generate deepfakes.
• Fake Media Layout Templates: Synthesized media combined with mockups of real news outlets like LA NACION.

Human Factors

• Deliberate Political Manipulation: Political actors intentionally spread lies to confuse the electorate.
• Troll Network Amplification: High-profile accounts actively reposted deepfakes to boost engagement.
• Voter Cognitive Vulnerability: Electorate struggled to quickly identify realistic AI-generated fakes.

Process and Methods

• Flawed Platform Moderation: X lacked automated, proactive detection for high-stakes political deepfakes.
• Reactive Enforcement Procedures: Removal of fake content required formal requests from prosecutors.

Regulatory Environment

• Electoral Blackout Loopholes: Regulations failed to prevent digital AI manipulation during the blackout.
• Weak Platform Regulation: Lack of immediate legal penalties for platforms hosting election deepfakes.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide clear, consistent details about the nature of the deepfakes, the platforms used for distribution, the political context, and the reactions of the affected parties. The role of AI as a tool for generating synthetic media is unambiguous.
• Ambiguities identified: The specific AI tools or models used to generate the deepfakes are not identified.

A coordinated political disinformation campaign in Argentina utilized AI-generated deepfakes of former President Macri and candidate Lospennato hours before the Buenos Aires legislative elections. The incident, aiming to mislead voters and manipulate democratic outcomes, represents a substantial threat to electoral integrity and information security, prompting judicial and regulatory intervention.

• Overall national security impact: Substantial
• Scope: Single nation
• Primary target: Argentina
• Alleged perpetrator: Accounts linked to the La Libertad Avanza party

Threat characteristics

• Imminence: Immediate. The deepfakes were deployed hours before the election, requiring rapid intervention from prosecutors and platforms to mitigate immediate electoral fraud.
• Autonomy: Human-controlled. The AI systems were used strictly as tools by human actors to generate synthetic media, with no autonomous decision-making or execution.
• Novelty: Evolved capability. While political deepfakes are an established concept, the targeted deployment of high-fidelity synthetic media in the final hours of an election represents an evolved capability.

Impact by dimension

• Physical security: Negligible. No physical threat, kinetic attacks, or critical infrastructure compromise was reported in connection with this incident.
• Information security: Substantial. Coordinated disinformation campaign using highly realistic AI deepfakes of former President Macri and candidate Lospennato to mislead voters hours before an election.
• Sovereignty: Substantial. Direct attempt to manipulate an electoral process and influence democratic outcomes in Buenos Aires, prompting investigations by the electoral tribunal and prosecutors.
• Economic security: Negligible. No significant economic manipulation, financial system attacks, or strategic technology theft occurred.
• Societal stability: Minor. Erodes public trust in digital media and democratic processes, but did not result in mass civil unrest or systemic human rights abuses.