This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A viral Facebook video featuring a deepfake of Citizen TV anchor Swaleh Mdoe was used to promote a fraudulent health product. The video falsely claimed a Kenyan doctor's home was bombed by pharmaceutical companies to suppress a 'miracle cure.' Investigations confirmed the footage was manipulated using AI to alter audio and lip movements, and the explosion footage was stolen from an unrelated incident in Ohio.
A purported AI-manipulated video falsely showing Citizen TV anchor Swaleh Mdoe reporting on the bombing of a Kenyan doctor's home circulated widely on Facebook. The video reportedly used AI-generated audio and visuals to fabricate a conspiracy in which pharmaceutical companies targeted the doctor for promoting a "miracle cure." In reality, the explosion footage was from Ohio, the doctor was fictitious, and the content aimed to manipulate viewers into purchasing an unproven health product.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to gain a personal advantage over others such as through cheating, fraud, scams, blackmail or targeted manipulation of beliefs or behavior. Examples include AI-facilitated plagiarism for research or education, impersonating a trusted or fake individual for illegitimate financial benefit, or creating humiliating or sexual imagery.
Also classified as
The video generated and spread false claims about a medical cure and a fabricated bombing to mislead viewers.
The deepfake video was created to deceive viewers into buying a fraudulent health product, exploiting their health concerns for financial gain.
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 incident was caused by human actors intentionally using deployed deepfake generation tools to create a fraudulent video to scam users.
Added categories peak at 3: Substantial (Epistemic Harm, Inferred)
A deepfake video depicting a Kenyan news anchor was used to promote a fraudulent health product. While the video garnered significant views and spread misinformation, it constitutes a financially motivated scam with negligible to minor national security implications, rather than a coordinated state-sponsored threat.
The AI system was used as a tool to perform a specific, well-defined function of manipulating video footage and generating synthetic voices based on direct inputs, without exhibiting autonomous planning or agency.
Limited Risk: The report describes 'deepfake technology' used to manipulate footage. Under the EU AI Act, AI-generated content such as deepfakes falls under Limited Risk, requiring transparency obligations to inform users.
The AI was used to manipulate existing video footage to alter lip movements and generate synthetic voices to match the targets.
The fact-checking report provides clear, verified details about the deepfake nature of the video, the sources of the stolen footage, and the fraudulent intent of the campaign. There is little ambiguity regarding the events or the role of AI.
None
Technology
Deepfake Voice and Lip Syncing
AI voice cloning and lip-syncing created highly realistic fake footage.
Algorithmic Recommendation
Platform algorithms promoted the video, leading to over 497,000 views.
Data Inputs
Stolen Video and Audio Footage
Original footage of news anchor and student doctor was used without consent.
Human Factors
Exploitation of Patient Fears
Ailing viewers easily trust miracle cures due to fear and health concerns.
Low Deepfake Awareness
Viewers struggled to identify subtle deepfake signs like lip-sync errors.
Process and Methods
Weak Platform Moderation
Facebook failed to detect and remove the deepfake video before it went viral.
Inadequate Deepfake Detection
Lack of robust automated tools to flag manipulated video and audio.
Management
Growth Over Safety Focus
Platform prioritized user engagement over proactive safety verification.
Regulatory Environment
Lack of Deepfake Policies
Absence of strict regulations governing AI-generated health misinformation.
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.
