This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Two women in Houston were targeted by scammers who used AI-generated deepfake videos of their likenesses to impersonate them on social media. The scammers gained access to the victims' accounts, posted fraudulent sales, and successfully solicited money from the victims' friends. The incident highlights the growing risk of AI-powered impersonation and social engineering.
Two Houston women reported being targeted in an alleged scam involving AI-generated videos that appeared to depict trusted friends. The purported deepfake avatars were reportedly used via social media and messaging apps to solicit access codes and promote fraudulent sales. Victims are reported to have lost control of their accounts, and friends of the impersonated individuals sent money believing the videos were authentic.
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.
Scammers used AI-generated deepfake videos and voice clones to impersonate trusted individuals and defraud their friends of money.
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 harm was caused by human scammers intentionally using deployed AI deepfake tools to defraud victims.
This incident involves localized consumer fraud in Houston, Texas, where scammers used AI-generated deepfakes to impersonate two women and steal minor sums of money from their friends. While demonstrating the growing accessibility of highly realistic AI impersonation tools, the event is a standard cybercrime case with negligible implications for national security.
The report describes scammers uploading a video or still image to an online AI tool and typing in a script for the avatar to say. This fits the 'Tool' definition as it performs a specific, well-defined function on direct input without autonomous planning.
Limited Risk: The report describes the use of AI-generated content such as deepfakes, which falls under Limited Risk due to transparency obligations.
The AI system was used to generate realistic deepfake videos and cloned voices of the victims to impersonate them.
The report clearly outlines the scam mechanism, the role of AI deepfakes, the impact on the victims, and the financial losses. While the specific AI tool used by the scammers is not named, the expert's demonstration and explanation make the technology's role unambiguous.
None. The incident is a straightforward case of AI-enabled impersonation fraud.
Technology
AI Avatar Generation
AI tools cloned faces and voices to create realistic avatars.
Advanced Lip-Synching
Still images were animated to match typed scripts convincingly.
Data Inputs
Publicly Available Media
Scammers harvested public videos and photos from social media.
Human Factors
Trust in Deepfake Identity
Victims believed the realistic video calls were their actual friends.
Sharing of Access Codes
Victims voluntarily sent MFA codes to scammers due to deception.
Process and Methods
Insecure Privacy Settings
Publicly accessible profiles allowed easy harvesting of media.
Management
Lack of Platform Response
Facebook failed to respond to reports or address the scam vector.
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.
