Purportedly AI-Generated Images and Videos Reportedly Spread Misinformation About Nicolás Maduro's Capture on X

Following the January 2026 U.S. raid and arrest of Venezuelan leader Nicolás Maduro, purportedly AI-generated images and videos circulated widely on platforms including X, reportedly depicting Maduro in fabricated scenarios. NewsGuard identified at least seven manipulated or synthetic visuals reaching over 14 million views, including AI-generated photos detected via Google Gemini's SynthID watermark.

Risk classification

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

The incident involves the coordinated and at-scale dissemination of AI-generated deepfakes and images to manipulate public opinion and manufacture consent regarding a geopolitical conflict.

Additional risk subdomains

• 3.1 False or misleading information: The incident involved the generation of highly realistic false information that misled millions of social media users.

Causal factors

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

The risk was caused by human actors intentionally generating and spreading AI-generated misinformation post-deployment to manipulate public opinion.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The reports describe the generation of deepfakes and AI-generated content, which under the EU AI Act are subject to transparency and labeling obligations to ensure users are aware they are interacting with or viewing AI-generated media.

AI system and alleged parties

• AI system: Gemini, Midjourney, Sora, unspecified (Google, OpenAI)
• AI purpose: Deepfake Video Generation; Deepfake Image Generation
• Behaviour type: Tool
• Alleged developer: xAI, Unknown image generator developers, Unknown deepfake technology developers, Google
• Alleged deployer: Unknown social media account operators, Unknown disinformation actors
• Alleged harmed parties: social media users, Nicolás Maduro, General public of Venezuela, General public, Epistemic integrity

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Severe. 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 Severe, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Minor, indirect Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Severe, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: The report explicitly describes the creation and spread of malicious AI-generated propaganda.

Directly caused: AI-generated videos of Venezuelan citizens falsely celebrating and images of Maduro in custody were spread widely, reaching over 14 million views on X.

Inferred additional harm: Additional unquantified reach across platforms like TikTok and Instagram where similar unlabeled AI content circulated.

Democracy

Reported: The report explicitly describes the use of AI to manufacture consent and manipulate public opinion.

Directly caused: The AI-generated content was used to manufacture support for U.S. military intervention and manipulate the narrative around the coup.

Inferred additional harm: The erosion of public trust in democratic discourse and international relations due to the difficulty of verifying authentic events.

Epistemic

Reported: The report explicitly describes a surge of visual misinformation blurring the line between fiction and reality.

Directly caused: AI-generated videos and images misled millions of social media users into believing Venezuelans were celebrating the U.S. intervention and that authentic custody photos of Maduro existed.

Inferred additional harm: Widespread erosion of shared truth and consensus reality as hyperrealistic AI content makes it harder for audiences to distinguish real from fake media.

People affected

• Occurrences reported: 1
• People reportedly harmed: 1
• People reportedly exposed: 14000000

Potential causes

Management

• Prioritizing Engagement Over Safety: Social media platforms allowed rapid viral spread before verifying content.

Technology

• Hyperrealistic Generative AI Tools: Tools like Sora and Midjourney make creating realistic fake videos easy.
• Inadequate Automated AI Detection: Detection and labeling tools yielded mixed results across major platforms.
• Imitation of Reality Improvements: AI is getting better at imitating reality, outpacing detection tech.

Data Inputs

• Lack of Media Fingerprinting: Absence of real media verification makes tracking authentic content hard.
• Use of Simple Text Prompts: Simple text prompts are used to generate hyperrealistic visuals easily.

Human Factors

• Confirmation Bias of Users: Netizens shared AI content to manufacture support for political narratives.
• High-Profile Amplification: Influential figures and politicians reshared unverified AI-generated slop.
• Viewer Inability to Spot Fakes: Hyperrealistic visuals blur the line between fiction and reality for users.

Process and Methods

• Slow Fact-Checking Mechanisms: Community Notes and fact-checkers react too slowly to stop viral spread.
• Inconsistent Platform Labeling: Some platforms labeled AI videos while others missed fabricated content.

Regulatory Environment

• Lack of Global Labeling Mandates: Inconsistent international laws fail to enforce AI content labeling.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide consistent, detailed accounts of the AI-generated misinformation campaign, including specific platforms (X, TikTok, Instagram), view counts (14 million), and the specific types of synthetic media generated. The role of AI tools in creating this content is explicitly stated and verified by fact-checking organizations like NewsGuard, BBC, and AFP.
• Ambiguities identified: The exact AI models used to generate some of the viral videos remain unconfirmed, though platforms like Sora and Midjourney are mentioned as representative tools.
• Alternative interpretations: None. The incident is clearly a case of politically motivated AI-generated disinformation.

Following the capture of Venezuelan leader Nicolas Maduro, a massive AI-generated disinformation campaign flooded social media, using hyperrealistic deepfakes to manufacture consent for foreign intervention and manipulate public opinion. Amplified by high-profile figures, the incident highlights the potent threat of generative AI as an information warfare tool during geopolitical crises.

• Overall national security impact: Substantial
• Scope: Multiple nations
• Primary target: Venezuela
• Other affected: United States
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Immediate. The rapid viral spread of deepfakes required immediate real-time response from fact-checkers and platforms to mitigate escalating public confusion.
• Autonomy: Human-controlled. AI systems were used strictly as tools by human actors to generate specific deepfake images and videos, without autonomous decision-making.
• Novelty: Evolved capability. While AI deepfakes in political conflicts are established, the scale, speed, and high-profile amplification represent a significant advancement in disinformation capabilities.

Impact by dimension

• Physical security: Negligible. No physical attacks, kinetic targeting, or critical infrastructure compromise were reported in connection with this incident.
• Information security: Substantial. Highly coordinated AI-generated disinformation campaign using deepfakes to manipulate public narratives and manufacture consent during a major geopolitical crisis, reaching millions.
• Sovereignty: Substantial. AI-generated content manufactured false narratives of public support for foreign military intervention and regime change, impacting international perceptions of national sovereignty.
• Economic security: Negligible. No significant threats to financial systems, strategic technology theft, or economic infrastructure were reported.
• Societal stability: Substantial. Widespread dissemination of unverified synthetic media eroded public trust and manipulated social cohesion at scale during a highly volatile political transition.