Child Predators Are Reportedly Generating Deepfake Nudes of Children to Extort Them

Internet Watch Foundation (IWF) has reported that it has found a manual on the dark web that encourages criminals to use "nudifying" AI tools to depict children naked in order to extort victims into providing graphic content.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.3 Fraud, scams, and targeted manipulation

The incident involves criminals using AI nudifying tools to generate non-consensual sexual imagery of children for targeted blackmail and extortion.

Additional risk subdomains

• 1.2 Exposure to toxic content: The AI tools are used to generate and expose victims to highly toxic child sexual abuse material.

Causal factors

• Entity: Human
• Intent: Intentional
• Timing: Post-deployment

The harm is caused by human criminals intentionally using deployed AI nudifying tools to blackmail children.

EU AI Act risk tier

• Risk tier: 1 Unacceptable

Unacceptable Risk: The report describes exploitative AI systems targeting vulnerabilities based on age (children) to facilitate blackmail and abuse.

AI system and alleged parties

• AI system: nudifying AI tools
• AI purpose: Deepfake Image Generation; Image Retouching
• Behaviour type: Tool
• Alleged developer: Unknown deepfake creators, Unknown AI developers
• Alleged deployer: Unknown deepfake creators, Unknown Dark Web users
• Alleged harmed parties: Children

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Substantial. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Substantial, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Minor
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Negligible, indirect Minor
• Psychological: direct Negligible, indirect Minor
• Epistemic: direct Negligible, indirect Negligible
• Child sexual exploitation and abuse: direct Substantial, indirect Substantial

Malicious content

Reported: The report explicitly describes the creation of 'nudified' non-consensual sexual imagery of children.

Directly caused: AI nudifying tools were used to directly generate non-consensual nude images of children from underwear photos.

Inferred additional harm: Widespread generation and potential dissemination of synthetic CSAM across dark web networks and online platforms.

Civil rights

Reported: The report explicitly describes severe violations of children's rights, including sexual exploitation and abuse.

Indirectly caused: The use of AI tools to generate CSAM and blackmail children violates their fundamental rights to safety, privacy, and protection from exploitation.

Inferred additional harm: Widespread violations of children's rights globally as these tools become more accessible to bad actors.

Privacy

Reported: The report explicitly describes privacy violations through the creation of nude images of people without their consent.

Indirectly caused: The generation of non-consensual synthetic nude images of children from their private photos violates their bodily privacy.

Inferred additional harm: Massive scale of non-consensual privacy violations against children and adults using easily accessible nudification apps.

Psychological

Reported: The report explicitly describes psychological harm in the form of blackmail and extortion of children.

Indirectly caused: Children (at least 2 reported) were blackmailed and extorted using AI-nudified images, causing severe distress, anxiety, and trauma.

Inferred additional harm: Hundreds of other children targeted by users of this dark web manual likely suffered severe psychological trauma and anxiety from extortion.

Child sexual exploitation and abuse

Reported: The report explicitly describes AI-generated CSAM and AI-facilitated extortion of children.

Directly caused: AI nudifying tools were used to generate synthetic CSAM (nude images of children).

Indirectly caused: The synthetic CSAM was used to blackmail children into producing and sending further real, graphic CSAM.

Inferred additional harm: Widespread proliferation of AI-generated CSAM and increased cases of self-generated CSAM coercion facilitated by AI tools.

People affected

• Occurrences reported: 1
• People reportedly harmed: 2
• People reportedly exposed: 2

Potential causes

Management

• Inadequate Platform Safeguards: Tech companies fail to implement strong safeguards to protect children.
• Lack of Platform Accountability: Social media platforms prioritize growth over robust child safety measures.

Technology

• AI Nudification Tools: Software allows users to easily remove clothing from images of children.
• Realistic Generative AI: AI generates astoundingly realistic child abuse content, easing exploitation.

Data Inputs

• Underwear Photos of Children: Perpetrators obtain underwear shots of children to feed into nudification AI.
• Self-Generated Child Images: Children are manipulated into sending photos that are used as AI inputs.

Human Factors

• Criminal Sharing of AI Manuals: Perpetrators share dark web manuals instructing others on AI exploitation.
• Lack of Parental Supervision: Young children use social media unsupervised, exposing them to risks.
• Child Vulnerability to Deception: Children are easily manipulated by opportunistic online predators.

Process and Methods

• Lack of AI Tool Safeguards: Nudification tools lack built-in blocks to prevent processing child images.
• Inadequate Content Filtering: Social platforms fail to block AI-generated abuse material effectively.

Regulatory Environment

• Lagging AI Regulations: Absence of active bans on harmful nudification tools facilitates their use.
• Delayed Safety Act Enforcement: The Online Safety Act duties of care are not yet fully operational.

Information quality

• Classification confidence: High
• Reason for confidence: The report from a reputable source (The Guardian) citing a major child protection charity (IWF) provides clear details on the existence of the dark web manual and the specific mechanism of AI-facilitated blackmail.
• Ambiguities identified: The exact AI models or platforms used are not named, and the precise number of victims is not quantified.
• Alternative interpretations: None.

An anonymous dark web manual detailing how to use AI nudifying tools to generate synthetic CSAM and extort children was discovered and reported to the UK National Crime Agency. While representing a severe violation of human rights and a growing challenge for law enforcement, the incident currently has minor direct national security implications and is managed within standard criminal justice frameworks.

• Overall national security impact: Minor
• Scope: Multiple nations
• Primary target: United Kingdom
• Other affected: Unknown
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Near-term. The proliferation of dark web manuals and accessible AI nudifying tools represents a rapidly developing threat requiring active law enforcement monitoring.
• Autonomy: Human-controlled. The AI tools are operated directly by human actors who make the final decisions to generate and use the synthetic media for extortion.
• Novelty: Evolved capability. While blackmail and CSAM are established threats, the integration of accessible AI nudification tools represents a significant technological evolution in how these crimes are committed.

Impact by dimension

• Physical security: Negligible. No physical systems, critical infrastructure, or kinetic military capabilities were targeted or affected by this incident.
• Information security: Negligible. No compromise of classified intelligence, state-sponsored disinformation campaigns, or espionage activities reported.
• Sovereignty: Negligible. Core government operations, electoral systems, and state authority remain unaffected, though legislative discussions on AI bans are occurring.
• Economic security: Negligible. No strategic technology theft, financial system attacks, or critical supply chain disruptions identified.
• Societal stability: Minor. The use of AI to generate synthetic CSAM and extort children represents a serious violation of human rights and privacy, currently managed under standard national law enforcement procedures by the NCA.