This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A blogger uncovered a scam operation where perpetrators used AI-generated faces to create a fake law firm website. The scammers used this fraudulent identity to send fake DMCA copyright infringement notices to website owners, attempting to coerce them into providing backlinks to improve the scammers' SEO rankings. The operation was dismantled after the blogger investigated the firm's credentials and confirmed the faces were synthetic.
GAN faces were allegedly used by scammers alongside a parked domain and a fake website to impersonate a Boston law firm.
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 perpetrators used GAN-generated synthetic faces to construct a fake law firm website and impersonate legal professionals to extort backlinks from website owners for SEO manipulation.
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 scam and extortion campaign were intentionally planned and executed by human actors who used GAN-generated faces as a tool to establish a fraudulent online identity.
Added categories peak at 2: Minor (Epistemic Harm, Inferred)
An online scam operation used GAN-generated synthetic faces to construct a fake law firm website and extort bloggers for SEO backlinks via fraudulent DMCA notices. The incident represents a minor national security concern, demonstrating an evolved capability of using synthetic media for social engineering, but was easily managed and lacked any state-actor involvement or critical infrastructure impact.
The GAN was used as a static tool to generate synthetic images of human faces on demand, which the human scammers then manually incorporated into their fraudulent website.
Limited Risk: The AI system was used to generate deepfake images (synthetic faces of non-existent people). Under the EU AI Act, such systems are subject to transparency obligations to ensure users are aware the content is AI-generated.
The GAN was used specifically to generate realistic, synthetic faces of non-existent people to populate the fake law firm's website.
The report provides a clear, first-hand investigative account of the scam, detailing how the GAN-generated faces were identified and how the fraudulent operation was verified and shut down.
None
Technology
Realistic GAN Face Generation
Generative AI creates highly realistic fake human faces.
Presence of Visual Artifacts
GAN faces contain subtle anomalies like distorted earrings.
Data Inputs
Large Facial Image Datasets
GANs require massive datasets of real faces to train effectively.
Human Factors
Inability to Detect GAN Faces
Users fail to recognize synthetic images without technical training.
Susceptibility to Social Engineering
People trust realistic-looking faces on professional websites.
Process and Methods
Abuse of Free GAN Tools
Scammers easily access free websites that generate synthetic faces.
Management
Lack of AI Ethics in SEO
Scammers exploit AI capabilities to optimize fraudulent SEO campaigns.
Regulatory Environment
Lack of Synthetic Media Regulation
No legal frameworks prevent or penalize using AI faces for fraud.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
