This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Multiple YouTube Live streams utilized AI-generated deepfake videos of Donald Trump to promote a fraudulent Bitcoin giveaway scam. The videos, which falsely depicted the former president at a NATO summit, instructed viewers to scan QR codes to send cryptocurrency. This incident is part of a broader trend of using AI tools to facilitate financial crimes.
Multiple YouTube Live streams reportedly featured a purported AI‑generated deepfake video depicting U.S. President Donald Trump at a NATO Summit press event. In the video, Trump reportedly appears to promote a Bitcoin "double-your-money" offer, directing viewers to scan an on‑screen QR code and send cryptocurrency to a specified address. The video was reportedly not authentic and is alleged to have been part of an ongoing YouTube-based crypto scam campaign.
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 video generated highly realistic but entirely fabricated statements from a public figure to mislead the public.
The incident involves the use of AI-generated deepfakes to impersonate a public figure for the purpose of executing a financial scam and defrauding cryptocurrency investors.
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 initiated by human scammers who intentionally deployed deepfake technology post-deployment to execute a financial scam.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
Scammers used AI-generated deepfakes of Donald Trump at a fake NATO summit to conduct a cryptocurrency giveaway scam on YouTube Live. While the incident demonstrates the increasing sophistication of synthetic media tools, it represents a minor national security threat focused on financial fraud rather than state-sponsored information warfare or critical infrastructure disruption.
The deepfake software is described as a tool used by scammers to generate realistic video and voice clones of Donald Trump based on direct inputs, without exhibiting autonomous planning or decision-making.
Limited Risk: The report describes the use of deepfake software to generate synthetic video and audio of a public figure. According to the EU AI Act definitions, AI-generated content such as deepfakes falls under Limited Risk (Level 3), requiring specific transparency obligations.
The AI system was used to generate a highly realistic deepfake video of Donald Trump with synchronized lip movements and a cloned voice to deceive viewers.
The report clearly details the nature of the scam, the platform used (YouTube Live), the target of the deepfake (Donald Trump), and the mechanism of the fraud (QR codes for Bitcoin). There is little ambiguity about the AI's role as a tool for generating deepfakes.
None. It is clearly a malicious deepfake scam.
Technology
Advanced Deepfake Software
AI tools made it easy to generate highly believable voice and video clones.
Convincing Lip-Syncing Tech
AI-generated speech matched the visual lip movements seamlessly.
Data Inputs
Real NATO Summit Footage
Scammers used authentic background visuals to mimic a real press event.
Malicious QR Codes
Visual overlays redirected victims to scam addresses.
Human Factors
Viewer Confusion Over Crypto
Followers were unsure which Trump-related crypto ventures were legitimate.
Plausibility of Endorsement
Trump's crypto-friendly stance made the fake giveaway seem plausible.
Process and Methods
Slow Video Suspension Process
Delays in taking down reported videos allowed scams to run live.
Inability to Block Duplicates
Duplicate versions of banned videos easily bypassed upload filters.
Management
Inadequate Live Stream Safety
Platform failed to address the persistent trend of celebrity deepfakes.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
