This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Scammers used AI-generated voice cloning and repurposed video footage to impersonate a Salt Lake City police chief in an email extortion attempt targeting a resident. The scam falsely claimed the recipient owed $100,000 to the federal government. Similar incidents involving AI-impersonated police officers have been reported in Tulsa and India, highlighting a growing trend of AI-enabled fraud.
A scammer used AI to create a deepfake video of Salt Lake City Police Chief Mike Brown, falsely claiming that a recipient owed $100,000 to the federal government. The video, sent via email from a fake SLCPD account, used a cloned voice and repurposed footage from a past interview.
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 voice-cloning technology to impersonate police officers and family members to commit targeted fraud and extortion.
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 was orchestrated by human actors who intentionally used deployed AI voice-cloning technology to create a fraudulent video for extortion.
Added categories peak at 2: Minor (Psychological Harm, Inferred)
Criminal actors leveraged AI-generated voice cloning and deepfake video technology to impersonate local police officials in extortion attempts targeting citizens in the US and India. While representing an evolution in fraud tactics and a minor threat to local public trust, the incident remains a localized criminal matter with negligible direct impact on national security.
The AI system was used as a tool by human scammers to perform a specific, well-defined function, namely cloning voices and generating synthetic audio from direct inputs. The reports state that 'scammers can create realistic voice recordings and even videos, known as voice cloning or deepfakes' using computer programs.
Risk Level 3. Limited Risk: The reports describe the use of 'AI-generated content such as deepfakes' and 'voice cloning' to impersonate individuals, which falls under the transparency obligations of Limited Risk systems.
The primary technology used in these scams was voice cloning to generate realistic audio of police officers and family members, combined with deepfake video generation in the Salt Lake City case.
The reports provide clear, consistent details about the nature of the scams, the technology used (voice cloning and deepfakes), and the specific targets. There is no conflicting information regarding how the AI was used.
None. The events are clearly identified as malicious impersonation scams.
Technology
Advanced Voice Cloning Tech
AI tools can realistically clone voices from short audio clips.
Deepfake Video Manipulation
Scammers synced cloned audio with old video footage.
Caller ID Spoofing
Scammers easily spoofed phone numbers to look like police.
Data Inputs
Public Media Exploitation
Scammers harvested old TV interviews from the internet.
Social Media Oversharing
Publicly shared personal audio and video provide training data.
Human Factors
Susceptibility to Urgency
Scammers exploit fear and urgency to bypass critical thinking.
Trust in Authority Figures
Victims comply when targeted by impersonated police chiefs.
Oversharing of Personal Data
People share details online, aiding customized scams.
Process and Methods
Weak Email Verification
Scammers used generic Google accounts instead of official domains.
Lack of Immediate Verification
Victims often act before verifying caller identity.
Spoofed Caller IDs
Scammers exploit telecom vulnerabilities to spoof police numbers.
Management
Rapid Tech Release
AI developers release powerful cloning tools with few safeguards.
Inadequate Media Protection
Organizations fail to protect official media from manipulation.
Lack of Public Education
Organizations fail to adequately warn citizens of AI-cloned scams.
Regulatory Environment
Unregulated AI Tool Access
Powerful voice-cloning tools are easily accessible online.
Irreversible Payment Methods
Scammers demand wire transfers or crypto, preventing recovery.
Lack of AI Watermarking
No enforced standards for watermarking generated synthetic audio.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
