This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
AI-generated deepfake videos and images are being used in a coordinated investment scam campaign on social media platforms like X and Facebook. These ads feature fabricated physical confrontations between public figures to lend credibility to fraudulent financial schemes. The Bank of England and other institutions have reported these scams, which exploit the public and bypass existing moderation systems.
Purported AI-generated ads on X reportedly depicted Bank of England governor Andrew Bailey and Reform UK leader Nigel Farage in fabricated physical confrontations on a BBC Question Time-style set. The ads allegedly used the staged Bailey-Farage clash to present an investment scam as legitimate BBC-style financial journalism.
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 deepfake videos depict public figures in violent physical confrontations, including holding a gun and kicking, exposing users to toxic and fabricated violent content.
The campaign generates and spreads false narratives and fake news articles about public figures to mislead the public.
The incident involves the use of AI-generated deepfakes of public figures to run fraudulent investment scams, which directly aligns with fraud, scams, and targeted manipulation under Domain 4.
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 is caused by human malicious actors who intentionally created and distributed deepfake advertisements post-deployment to defraud the public.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
A coordinated global financial fraud campaign utilizing sophisticated AI-generated deepfakes of central bank governors and political figures to lure victims into investment scams. Suspected to be run by a Russian-language criminal network, the incident highlights the evolving threat of synthetic media to institutional trust and economic security, though it does not pose an imminent threat to physical infrastructure.
The AI systems were used as tools by scammers to generate synthetic deepfake videos and images of public figures. The systems performed specific, well-defined functions of image and video generation based on direct inputs, without exhibiting independent planning or autonomous decision-making.
Limited Risk: The report describes the use of AI-generated content such as deepfakes, which under the EU AI Act are classified as limited risk and subject to specific transparency obligations to ensure users are informed of their synthetic nature.
The AI systems were primarily used to generate highly realistic but entirely fabricated videos and images of public figures to deceive the public.
The report clearly details the nature of the deepfake scam, the platforms involved, the specific public figures targeted, and the fraudulent intent. There is no ambiguity about the fact that these are AI-generated deepfakes used for financial scams.
None. The evidence clearly points to a financially motivated cyber-scam using synthetic media.
Technology
Convincing Deepfake AI
AI tools enabled highly realistic fake videos of public figures.
Ad Security Evasion
Scammers bypassed automated security filters on social platforms.
Data Inputs
Misuse of Likeness Data
Scammers used images of public figures without authorization.
Human Factors
Vulnerability to Scams
Vulnerable online users are easily exploited by deceptive ads.
Over-reliance on User Reports
Platform relies on users to report scams rather than proactive detection.
Process and Methods
Slow Content Moderation
Fake ads remain online for days before being successfully removed.
Flawed Ad Review Process
Inadequate screening allowed fraudulent deepfake ads to go live.
Management
Drastic Safety Staff Cuts
X gutted safety teams by eighty percent, reducing moderation capacity.
Regulatory Environment
Weak Regulatory Enforcement
Watchdogs lack the power to enforce immediate removal of fake ads.
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.
