This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Scammers are using AI-generated deepfakes of Princess Leonor of Spain on TikTok to conduct a financial fraud scheme targeting vulnerable individuals in Latin America. The perpetrators impersonate the princess or her legal representatives to promise large financial prizes, then coerce victims into paying multiple 'fees' or 'taxes' to release the funds. The scheme exploits TikTok's algorithm to gain visibility and uses AI to create convincing video content, resulting in significant financial loss for victims.
Scammers have allegedly been using deepfake technology and fake social media accounts to reportedly impersonate Leonor, Princess of Asturias, targeting vulnerable individuals in Latin America. Victims were reportedly lured with promises of financial aid, requiring payments for "fees" or "taxes" before receiving funds, only to allegedly be defrauded repeatedly. TikTok accounts with thousands of followers reportedly amplified the scheme using AI to increase the appearance of credibility.
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.
The incident involves scammers using AI-generated deepfakes to impersonate a royal figure and defraud specific vulnerable victims 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 financial scam was intentionally orchestrated by human actors using deployed AI generation tools to create deepfakes.
Added categories peak at 3: Substantial (Epistemic Harm, Inferred)
An international financial fraud scheme utilized AI-generated deepfakes of Princess Leonor of Spain to target vulnerable individuals in Latin America. While the incident demonstrates the evolving threat of synthetic media in facilitating cross-border crime, its national security impact remains minor as it lacks geopolitical motivation or systemic infrastructure targets.
The reports describe scammers using AI tools to generate deepfake videos and voice clones of Princess Leonor. The AI system performed specific, well-defined functions of image and voice generation based on direct inputs from the scammers, acting as a tool.
Limited Risk. The report describes the use of 'AI tools to mimic Leonor's voice and likeness' and 'a fake Princess Leonor — created using AI tools to mimic her movement and speech', which constitutes 'AI-generated content such as deepfakes'.
The AI was primarily used to generate realistic deepfake videos of Princess Leonor to deceive victims, along with voice generation to mimic her speech.
The reports provide consistent, detailed accounts of the scam, including a specific victim interview, the platforms used, the role of AI deepfakes, and the response from authorities and TikTok. There is high certainty regarding the classification of the incident as a financial scam utilizing AI deepfakes.
None. The incident is clearly a coordinated financial fraud campaign using impersonation and deepfakes.
Technology
AI Voice and Likeness Mimicry
AI tools used to mimic Princess Leonor's voice and movement to deceive victims.
Engagement-Driven Algorithms
TikTok's algorithm amplified fake accounts due to high user engagement.
Data Inputs
Lack of Official Royal Accounts
Absence of real royal profiles allowed scammers to fill the identity vacuum.
Manipulated Profile Images
Scammers added fake blue verification checks to profile pictures to gain trust.
Human Factors
Trust in Royal Authority
Victims trusted the royal identity, making them susceptible to the scam.
Vulnerability of Target Audience
Elderly and financially vulnerable individuals were targeted for exploitation.
Process and Methods
Slow Content Moderation
TikTok was slow to remove reported fake accounts, letting some stay active.
Inadequate Account Verification
Lack of robust verification allowed fake accounts to mimic verified profiles.
Management
Failure to Proactively Monitor
TikTok relied on external reports rather than proactively detecting royal fakes.
Regulatory Environment
Lack of Cross-Border Enforcement
Scammers operated from Dominican Republic, hindering local law enforcement.
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.
