How Are Malicious AI Models Fueling Explosive Cybercrime on the Darknet?

Can Your Business Survive the Alarming Rise of Darknet AI Threats?

Artificial intelligence (AI) is rapidly transforming both legitimate business operations and the cybercrime landscape. The latest Check Point AI Security Report 2025 reveals a surge in the use of AI by cybercriminals, who are weaponizing advanced technologies to launch more sophisticated, scalable, and damaging attacks than ever before.

Key Findings from the AI Security Report 2025

Widespread AI Adoption and Data Risks

• 51% of organizations now use AI services monthly.
• Alarming statistic: 1 in 80 AI prompts within corporate networks contains highly sensitive data, posing a significant risk of data leakage.

Criminal AI-as-a-Service on the Dark Web

• AI-powered tools for cybercrime, such as GoMailPro (integrated with ChatGPT), are sold for $500/month.
• Fraudulent AI-based phone services can cost up to $20,000 or are available for a base price plus per-minute fees.
• These services democratize access to advanced cyberattack tools, lowering the barrier for entry for would-be attackers.

Surge in AI Account Trading

• Stolen credentials for popular AI platforms (like ChatGPT) are traded on the dark web, enabling anonymous creation of malicious content and evasion of platform restrictions.

AI-Powered Malware and Automated Attacks

• Groups such as FunkSec use AI in at least 20% of their operations to develop malware and analyze stolen data, increasing the speed and efficiency of attacks.
• AI-driven data mining tools organize and validate stolen credentials, making cybercrime more profitable and precise.

Disinformation, Deepfakes, and Data Poisoning

• Disinformation campaigns have reached new heights, with one network producing 3.6 million articles in a year to manipulate AI systems.
• Around 33% of queries to leading Western AI systems were found to contain this manipulated content.
• Deepfake technology enables real-time impersonations, making social engineering attacks more convincing and harder to detect.
• LLM (Large Language Model) poisoning is a growing threat, where attackers corrupt training data to inject malicious outputs into AI models, despite strict validation efforts by leading vendors.

The Rise of Dark LLMs and Digital Twins

Dark LLMs

Maliciously modified AI models such as FraudGPT, WormGPT, and GhostGPT are specifically designed to bypass safety controls. These are sold on darknet forums, often with subscription models and user support, enabling scalable cybercrime.

Digital Twins

AI-driven replicas can convincingly mimic human behavior, speech, and even thought patterns, making traditional identity verification systems increasingly ineffective.

Why This Matters for Businesses

Escalating Threats

Cybercriminals are leveraging AI at every stage of their operations, from generating phishing emails and deepfake videos to automating malware creation and data mining.

Data Loss and Compliance Risks

With 7.5% of AI prompts containing potentially sensitive information, organizations face critical challenges in protecting data integrity and ensuring compliance.

Erosion of Digital Trust

As AI-generated content becomes indistinguishable from authentic material, the very foundation of digital trust is at risk.

Recommended Defensive Strategies

• Adopt AI-Assisted Threat Detection: Use advanced AI to identify and counteract AI-driven threats in real time.
• Enhance Identity Verification: Move beyond traditional methods by incorporating multi-layered, AI-powered verification systems.
• Strengthen Threat Intelligence: Integrate AI context into your threat intelligence to stay ahead of evolving tactics.
• Educate and Train Staff: Regularly update employees on the latest AI-driven attack vectors and how to recognize them.

Summary

The weaponization of AI by cybercriminals is no longer a distant threat-it is reshaping the digital landscape today. From Dark LLMs and deepfakes to automated malware and large-scale disinformation, attackers are exploiting AI faster than many organizations can defend against them. To protect sensitive data, maintain compliance, and preserve digital trust, businesses must urgently update their cybersecurity strategies, embracing AI both as a tool for defense and a lens through which to view all emerging threats.