Cybersecurity
AI is making it easier for small-time criminals to launch their own ransomware operations
Cybercriminals are increasingly turning to AI tools to automate phishing, code review, and extortion tactics, democratizing an attack method that once required membership in established criminal networks.
The threat of ransomware and data extortion has never been more acute. While fully autonomous AI-powered attacks haven't yet materialized, criminal groups are weaponizing large language models to streamline the manual labor of their operations—making it possible for less experienced actors to compete with established operations.
The scale is accelerating#
Tracking exact attack volumes is challenging because victims often stay silent and threat actors exaggerate, but security researchers have documented a troubling upward trend. Claims of ransomware incidents on leak sites—where criminal groups boast about their targets—have spiked dramatically over recent months, with estimates suggesting as many as 3,000 attacks during summer alone. That represents a potential 20 to 40 percent jump from spring figures. More concerning is the shift in geography: while US targets remain under heavy assault, international incidents have surged noticeably. According to Cynthia Kaiser, a former FBI cyber deputy director now at Halcyon, "Internationally we're seeing a lot more attacks. That to me is the really new part, and I do think it is in part fueled by AI."
How AI is changing the attack playbook#
Cybercriminals aren't deploying fully sentient ransomware bots—the traditional extortion model already works too well to abandon. Instead, they're using AI to automate the grunt work that previously required human expertise:
- Phishing sophistication: LLMs generate more convincing deceptive emails tailored to victims' languages and contexts. In Japan, for instance, AI-generated content in native Japanese has suddenly made scams far more convincing to local targets.
- Negotiation: AI assists in drafting ransom demands and managing victim communications.
- Code validation: Attackers use AI to review their exploit code and identify bugs.
- Lateral movement: Criminals leverage AI to compromise remote-access tools from rival groups, creating backup pathways into victim networks.
The democratization of ransomware#
Traditionally, cybercriminals relied on "ransomware-as-a-service" models, functioning like criminal franchises where operators paid established groups a cut of profits. AI has upended that ecosystem. Researchers now observe a proliferation of lone-wolf and small-team operations, many run by inexperienced actors who would have struggled to operate independently just years ago. ReliaQuest documented one new group, calling itself Global Secret Group, claiming 48 breaches in recent months. Another, dubbed Storm, listed 71 alleged victims in just 20 days. A particularly striking case revealed an operator running a new ransomware group as an inexperienced teenager who claimed over $800,000 in just under two weeks.

