AI-Generated Images Purporting to Show Donald Trump Kissing White House Aide Natalie Harp Went Viral on Social Media

AI-generated images purporting to show U.S. President Donald Trump kissing White House aide Natalie Harp on a golf course spread across social media; one X post received over 10 million views. Forensic analysis and provenance checks indicated the circulating images were synthetic, and the White House called them fake. The original creator remained unidentified.

Risk classification

• Primary risk domain: 3 Misinformation
• Primary risk subdomain: 3.1 False or misleading information

The incident involves the generation and viral spread of discrete false and misleading AI-generated images depicting a fabricated political scandal.

Additional risk subdomains

• 4.3 Fraud, scams, and targeted manipulation: The deepfake images targeted specific public figures, Donald Trump and Natalie Harp, creating a false and potentially damaging narrative about their personal relationship.

Causal factors

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

The incident was caused by human actors intentionally generating and sharing deepfake images of public figures to mislead or entertain social media users post-deployment.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The incident involves AI-generated content such as deepfakes, which are subject to transparency and watermarking obligations under the EU AI Act.

AI system and alleged parties

• AI system: ChatGPT, Gemini (Google, OpenAI)
• AI purpose: Deepfake Image Generation; Image Retouching
• Behaviour type: Tool
• Alleged developer: Synthetic media generation technology developers, Synthetic image generation technology developers, OpenAI, Large language model developers, Deepfake technology developers, Chatbot developers
• Alleged deployer: Synthetic media creators, Mykhailo Rohoza, Keith Edwards, Information manipulation actors, Deepfake creators
• Alleged harmed parties: Natalie Harp, General public misled by deepfake content, General public, Epistemic integrity, Donald Trump

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

Malicious content

Reported: The report explicitly describes the creation and spread of misleading synthetic media on social platforms.

Directly caused: AI-generated deepfake images of Donald Trump and Natalie Harp were created and spread across platforms like X, YouTube, and Threads.

Democracy

Reported: The report describes concerns regarding the erosion of public trust in the visual record due to the spread of AI fakes.

Indirectly caused: Experts noted that sowing distrust in the visual record leads to a situation where 'nothing is real' and 'nothing's fake,' undermining the shared information ecosystem necessary for democratic discourse.

Inferred additional harm: The viral spread of political deepfakes close to elections can erode trust in democratic institutions and political figures by fabricating scandals.

Epistemic

Reported: The report explicitly describes epistemic harm through the viral spread of fabricated images that misled some online users.

Directly caused: Fake images purporting to show Donald Trump kissing Natalie Harp amassed millions of views, fooling some social media users and requiring extensive fact-checking efforts to debunk.

Indirectly caused: Fact-checkers and journalists had to spend significant time and resources to verify and deconstruct the hoax, making content verification a complex and lengthy process.

Inferred additional harm: The incident contributes to a broader erosion of shared truth, making it harder for the public to distinguish between real and synthetic media.

People affected

• Occurrences reported: 1
• People reportedly harmed: 2
• People reportedly exposed: 10000000

Potential causes

Technology

• Generative AI Image Synthesis: Generative AI tools allowed the creation of realistic fake political images.
• Anatomical Generation Errors: AI image models struggle with hands, generating a hand with six fingers.
• Watermark Removal Vulnerability: Blurring or degrading images could bypass standard AI detection watermarks.

Data Inputs

• Low-Quality Source Images: Degraded images bypassed initial automated AI detection systems.

Human Factors

• Lack of Source Verification: Users and creators shared the image without verifying its original source.
• Intentional AI Image Enhancement: A user used ChatGPT to enhance and brighten the blurry fake image.
• Deliberate Sharing for Engagement: Influencers used the fake image as thumbnails to attract video views.

Process and Methods

• Complex Fact-Checking Process: Verifying AI content has become a complex and time-consuming manual process.
• Inadequate Thumbnail Screening: Platforms failed to screen or flag AI-generated thumbnails before publication.

Information quality

• Classification confidence: High
• Reason for confidence: The report provides clear, detailed, and verified facts about the origin, spread, and detection of the deepfake images, including statements from the platforms, experts, and individuals involved in sharing the content.
• Ambiguities identified: The original creator of the deepfake image and the exact tool used to generate the initial high-resolution version remain unknown.
• Alternative interpretations: None, the event is clearly a political deepfake misinformation incident.

An AI-generated deepfake depicting Donald Trump went viral on social media, highlighting ongoing challenges in digital content verification. The incident represents a minor national security concern, characterized as an established information-integrity threat with low coordination and no physical, sovereign, or economic disruption.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. The immediate viral event has been resolved and debunked, representing an ongoing, long-term strategic concern regarding synthetic media verification.
• Autonomy: Human-controlled. The generative AI tools and ChatGPT were entirely directed and manipulated by human users to create and edit the images.
• Novelty: Established threat. Political deepfakes and manipulated images of high-profile leaders are an established threat with numerous prior occurrences.

Impact by dimension

• Physical security: Negligible. The incident involves purely digital, synthetic media of a political nature and has no physical security or critical infrastructure implications.
• Information security: Minor. Represents a low-level, uncoordinated information operation. While the deepfake went viral with over 10 million views, it was quickly debunked by fact-checkers and lacked evidence of a sophisticated, state-sponsored foreign intelligence campaign.
• Sovereignty: Minor. The deepfake targeted a political figure and former President, but did not disrupt core government operations, electoral infrastructure, or constitutional processes.
• Economic security: Negligible. No impact on financial systems, critical supply chains, or national technological competitive advantage.
• Societal stability: Minor. Contributes to the erosion of public trust in visual media and digital discourse, but did not trigger civil unrest, mass surveillance, or systemic human rights violations.