This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A GOP pollster circulated AI-generated images depicting Black men supporting the Republican Party to create a false impression of racial diversity within the party. The incident highlights concerns regarding the use of synthetic media to distort political representation and target specific demographics with disinformation. The report also notes broader trends of AI being used to generate misleading content involving Black celebrities and political figures.
Purportedly AI-generated images circulated by GOP pollster Patrick Ruffini reportedly depicted Black men appearing to support the Republican Party shortly before Black History Month. The allegedly fabricated photos created a misleading impression of racial diversity within the GOP and may have shaped public perception. The episode raises concerns about the reported use of AI to distort political representation and the potential impact on Black communities.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to conduct large-scale disinformation campaigns, malicious surveillance, or targeted and sophisticated automated censorship and propaganda, with the aim of manipulating political processes, public opinion, and behavior.
Also classified as
The AI-generated images falsely represented Black voters as props to create a misleading illusion of racial diversity within a political party.
The incident involves a political actor using AI-generated synthetic media to conduct targeted political influence and spread misleading representations to manipulate public opinion.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The risk was caused by a human pollster intentionally choosing to generate and distribute fake AI images to influence public perception.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
A domestic political pollster used AI-generated images to falsely depict Black support for a political party, highlighting the growing use of synthetic media in political campaigns. While easily detected and debunked, the incident underscores the evolving challenge of domestic disinformation targeting specific demographics.
The AI system was used as a simple image generation tool to produce specific synthetic images based on direct user prompts, without exhibiting autonomous planning or agency.
Limited Risk: The incident involves the creation of AI-generated synthetic media (deepfakes), which falls under transparency and disclosure obligations to ensure users are aware the content is AI-generated.
The AI system was used to generate realistic but entirely fabricated images of non-existent Black supporters to be shared on social media.
The report provides clear details about the specific actor, the platform used for distribution, the nature of the generated images, and the public reaction. While the exact AI model used is not specified, this does not impede the classification of the risk or harm.
None, the political and representational nature of the incident is clear.
Technology
AI Generation Artifacts
AI generated anomalous details like a third arm and weird flag images.
AI Media Manipulation
Creators used AI to generate fake and manipulated media of celebrities.
Human Factors
Deceptive Political Motivation
Pollsters used fake AI images to create an illusion of racial diversity.
Malicious Creator Intent
YouTube creators targeted Black celebrities with AI-generated disinformation.
Process and Methods
Inadequate Content Filtering
Platforms allowed AI-manipulated salacious content to flood and spread.
Management
Deceptive Outreach Practices
Prioritizing the illusion of diversity over genuine voter engagement.
Regulatory Environment
Absence of AI Regulation
Lack of active legislation protecting voters from AI-driven disinformation.
AI system involved
An AI system was involved, but the reports do not identify it.
Identified by automated extraction; can differ from the alleged developers above.
