Group-IB: AI fuels fifth wave of industrial cybercrime

Infosecurity Magazine reports that Group-IB has declared that AI is powering a "fifth wave" of cybercrime, dramatically industrializing attacks by making sophisticated tools cheap, scalable, and accessible to criminals of all skill levels.

In a new report, the company details how generative AI is being weaponized to create "synthetic identity kits" with deepfake video actors and cloned voices for as little as $5, enabling fraud and bypassing authentication systems. Dark web discussions on AI criminal tools have surged from under 50,000 messages annually pre-2022 to approximately 300,000 per year since 2023. Phishing has entered an "agentic AI" era, with kits automating victim targeting, lure generation, and campaign adaptation for monthly fees comparable to a Netflix subscription.

Furthermore, threat actors are moving beyond abusing public chatbots to developing proprietary "dark LLMs" like Nytheon AI, unrestricted, self-hosted models fine-tuned for malware development and scams, available via subscriptions. Group-IB's CEO Dmitry Volkov stated AI turns "human skills into scalable services," while former Interpol cybercrime director Craig Jones noted it hasn't created new motives but has "dramatically increased the speed, scale and sophistication" of attacks, marking a fundamental shift in the threat landscape.