This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
The FBI issued a warning regarding a surge in sextortion cases where malicious actors use AI-generated deepfakes to harass and extort victims. By manipulating benign photos or videos sourced from social media, attackers create realistic sexually explicit content to coerce victims into paying ransoms or providing further illicit material. This trend affects both minors and adults, posing significant challenges for victims in removing the content once it is circulated online.
The FBI reported an increase in sextortion cases featuring the use of fake, including AI-generated, images or videos created from content posted on their social media sites or web postings, provided to the malicious actor upon request, or captured during video chats.
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 AI-generated sexually explicit deepfakes are circulated on public forums, social media, and pornographic websites, exposing victims and online users to highly toxic and non-consensual sexual content.
The incident involves malicious actors using AI deepfake technology to target specific individuals for blackmail, extortion, and targeted harassment to obtain financial payment or illicit content.
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 sextortion schemes and harassment are driven by the intentional decisions and actions of malicious human actors utilizing deployed AI deepfake generation tools.
Added categories peak at 3: Substantial (Child Sexual Exploitation & Abuse (CSEA), Indirect)
The FBI issued a public warning regarding scammers using AI deepfake tools to generate realistic sexually explicit videos from single benign photos for sextortion. This evolved threat primarily impacts societal stability and individual privacy, particularly targeting minors and non-consenting adults, requiring active intervention from federal law enforcement.
The report describes deepfake software and cloud-based services as tools that users run to perform a specific function: manipulating a benign photo into a sexually explicit video. The AI operates on direct user input without context retention or autonomous decision-making, fitting the definition of a Tool.
Risk Level 3. Limited Risk: The report describes AI-generated content such as deepfakes, which are explicitly classified under Limited Risk due to transparency obligations to ensure users are aware of the AI's synthetic nature.
The AI systems are explicitly used to generate realistic synthetic videos and images ('deepfakes') by manipulating benign photographs of victims.
The FBI warnings provide clear, authoritative, and consistent information regarding the mechanics of the sextortion schemes, the role of AI deepfake technology, and the nature of the harms caused to victims. While specific statistics on the exact number of victims or financial losses are absent, the overall risk profile and taxonomic classifications are highly clear and unambiguous.
None. The reports clearly describe a malicious human-driven extortion scheme leveraging AI tools.
Technology
Advanced Deepfake Software
AI tools create realistic videos using only a single image of a face.
Easily Bypassed Guardrails
Built-in safety checks to prevent abuse are easy to skirt in practice.
Underground Market Tools
Unregulated deepfake services are available in underground markets.
Data Inputs
Public Social Media Photos
Scammers harvest benign photos posted on public social media accounts.
Captured Video Chat Media
Malicious actors record and capture images during video conversations.
Unrestricted Web Postings
Personal images are easily copied from open web postings and search.
Human Factors
Oversharing Personal Media
Users post personal photos and videos online without realizing the risks.
Accepting Unknown Requests
Users accept friend requests and communicate with unfamiliar individuals.
Weak Account Security
Lack of complex passwords and MFA allows accounts to be easily hacked.
Process and Methods
Difficult Content Removal
Victims face major challenges in removing circulated content from the web.
Inadequate Content Filtering
Platforms struggle to detect and block AI-generated explicit content.
Unmonitored Image Sharing
Direct messaging and public forums allow easy sharing of fake media.
Management
Lack of Proactive Protections
Platform management fails to restrict public access to user photos.
Inadequate User Education
Platforms do not adequately warn users about deepfake exploitation risks.
Lack of Privacy by Default
Default platform settings leave user profiles and friend lists public.
Regulatory Environment
Difficult Statute Enforcement
Enforcing federal criminal statutes is hard across digital platforms.
Inadequate Cyber Laws
Existing legal frameworks struggle to keep pace with rapid AI progress.
Lack of Platform Liability
Regulations do not hold platforms liable for hosting deepfake content.
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.
