ProKYC Tool Allegedly Facilitates Deepfake-Based Account Fraud on Cryptocurrency Exchanges

Cato CTRL security researchers reported that the cybercriminal group ProKYC is selling a deepfake tool capable of bypassing biometric and two-factor authentication (2FA) systems on cryptocurrency exchanges. The tool creates synthetic identities using AI-generated videos and forged documents, enabling fraudulent account creation. A demo video from ProKYC shows the tool in action against ByBit, allowing attackers to verify fake accounts for purposes such as money laundering and identity theft.

Risk classification

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

The ProKYC tool is designed and sold specifically to facilitate financial fraud, scams, and identity deception by bypassing KYC security controls on cryptocurrency exchanges.

Additional risk subdomains

• 2.2 AI system security vulnerabilities and attacks: The tool exploits vulnerabilities in biometric facial recognition and document verification systems to inject synthetic media and bypass security controls.

Causal factors

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

The risk is caused by human threat actors intentionally developing, selling, and deploying a deepfake tool to commit financial fraud after the underlying AI models have been trained.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Risk Level 3: The system is a deepfake generation tool used to create synthetic videos and documents, which falls under Limited Risk due to transparency obligations for AI-generated content.

AI system and alleged parties

• AI system: ProKYC deepfake tool
• AI purpose: Deepfake Video Generation; Deepfake Image Generation
• Behaviour type: Tool
• Alleged developer: ProKYC
• Alleged deployer: ProKYC
• Alleged harmed parties: Cryptocurrency investors, Cryptocurrency exchanges, ByBit

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Minor. 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 Minor, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Minor, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: Yes, the report describes the creation of deepfake videos and forged documents to bypass identity verification.

Directly caused: The tool directly generates high-quality forged documents, such as a fake Australian passport, and synthetic videos of fake personas to pass facial recognition challenges.

Inferred additional harm: It is likely that numerous fake documents and deepfake videos have been generated and uploaded to various cryptocurrency exchanges by buyers of the tool.

Epistemic

Reported: Yes, the report describes the creation of highly realistic fake documents and synthetic videos to deceive authentication systems.

Directly caused: The tool fabricates realistic identity documents and synthetic video streams to deceive biometric verification systems into believing a real person is presenting a valid ID.

Inferred additional harm: The proliferation of such tools undermines the reliability of digital identity verification and erodes trust in remote authentication mechanisms.

People affected

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

Potential causes

Management

• Prioritizing Frictionless UX: Management prioritizes easy onboarding over strict biometric security.
• Poor Threat Intelligence: Firms fail to track emerging dark web threats like ProKYC.

Technology

• Camera Input Spoofing: Tool injects pre-recorded deepfake video directly into the camera stream.
• Lax Facial Recognition: Algorithms ignore minor glitches to avoid false positives for users.
• Accessible Generative AI: Commodity AI tools make high-quality face generation extremely easy.

Data Inputs

• Synthetic Biometric Data: Deepfake videos mimic required head movements to trick liveness checks.
• High-Quality Document Forgery: AI-generated photos are layered onto realistic passport templates.

Human Factors

• Lack of Manual Oversight: Operators do not manually inspect suspicious verification attempts.

Process and Methods

• Weak Liveness Detection: Verification processes fail to detect virtual camera injection.
• Inadequate Glitch Detection: Processes do not flag digital artifacts and eye movement anomalies.

Regulatory Environment

• Lax Biometric Regulations: Regulatory standards do not mandate advanced anti-deepfake checks.

Information quality

• Classification confidence: High
• Reason for confidence: The report clearly details the functionality of the ProKYC tool, its target (cryptocurrency exchanges like ByBit), and how it operates to bypass 2FA. There is little ambiguity about its purpose and capabilities.
• Ambiguities identified: The report does not specify the exact underlying AI models used by ProKYC, nor does it quantify the exact number of successful frauds committed using this specific tool.

The ProKYC deepfake tool represents an evolved cybercriminal capability designed to bypass biometric and document-based identity verification on financial platforms. While it poses a minor threat to financial and economic security by enabling money laundering and fraud, it does not currently present a direct threat to physical infrastructure or state sovereignty.

• Overall national security impact: Minor
• Scope: Multiple nations
• Primary target: No clear primary
• Other affected: Multiple nations
• Alleged perpetrator: ProKYC

Threat characteristics

• Imminence: Long-term. Represents an ongoing capability and strategic concern regarding the erosion of biometric verification rather than an active, immediate national crisis.
• Autonomy: Human-controlled. The tool generates media based on direct inputs and is actively deployed by human threat actors to bypass security.
• Novelty: Evolved capability. Deepfakes and document forgery are established, but integrating virtual camera injection to bypass liveness checks represents an evolved capability.

Impact by dimension

• Physical security: Negligible. The incident is strictly digital financial fraud on cryptocurrency exchanges with no physical or critical infrastructure implications.
• Information security: Minor. Uses synthetic media and forged documents, but the target is commercial financial institutions for fraud rather than state-sponsored intelligence or information warfare.
• Sovereignty: Minor. The tool generates forged government documents like passports, which mimics state-issued credentials, but its use is aimed at commercial KYC bypass rather than state sovereignty.
• Economic security: Minor. Facilitates financial fraud and money laundering on cryptocurrency exchanges, posing a minor threat to financial system integrity and remote identity verification mechanisms.
• Societal stability: Negligible. Limits itself to financial fraud via synthetic personas, with minimal direct impact on civil liberties, mass surveillance, or societal stability.