ToolNavs Find Useful AI Tools
Submit Sign in
Back to AI information
AI Ransomware Hits Its Stride: Claims Jumped 30% in Q3 2026

AI Ransomware Hits Its Stride: Claims Jumped 30% in Q3 2026

AI information • Admin • • 25 views

On November 14, 2026, WIRED reported in its Kernel Panic column that AI is turning ransomware into a lower-barrier business: over the summer of 2026, ransomware groups posted nearly 3,000 alleged victim claims on leak sites, up 20% to 40% from the spring, with the growth driven mainly by attacks outside the United States. Researchers believe AI is at least partly behind the surge, though a fully AI-run ransomware attack has yet to appear.

Claims Are Up, Confirmations Are Not

The numbers need to be taken apart. Three trackers count differently, yet point the same way: Halcyon logged 2,766 claims in the third quarter, up 39.1% from the previous quarter, spanning 111 groups and 122 countries, with more than 1,000 claims in August alone — its highest monthly total on record. ReliaQuest counted roughly 2,928, up 30% from 2,252 in the second quarter. Comparitech counted 2,627, up 29% quarter on quarter and 61% year on year. But a claim is not a confirmed attack: of Comparitech's 2,627, only 247 were confirmed by the victims themselves, or 9.4%. Almost all the growth came from new faces: 43 first-time groups accounted for 93% of the quarter's increase, while groups already active in the second quarter added barely 1%.

AI Isn't Running the Extortion Itself — It's Tearing Down the Entry Barrier

AI is not changing extortion itself; it is changing how easy it is to get in. ReliaQuest's analysis shows the new group Global Secret Group claimed 48 victims, more than half of them listed on a single day, while another newcomer, Storm, listed 71 victims in 20 days. Brandon Tirado, ReliaQuest's director of threat research, said the growth reflects the acceleration AI provides, letting attackers strike out on their own instead of relying on traditional ransomware-as-a-service platforms; his team recently identified one new group as the work of an inexperienced teenager who made more than $800,000 in under two weeks. Cynthia Kaiser, Halcyon's research lead and a former FBI cyber deputy director, also believes AI is partly fueling the rise.

For now, AI does the grunt work in the ransomware chain: writing more convincing phishing emails, researching targets, checking code, and handling ransom negotiations. Allan Liska, a longtime ransomware researcher at Recorded Future, notes that successful groups are still largely human-run, and fully automated ransomware has yet to emerge — partly because the existing criminal business model already pays well. But Halcyon's report hides a more alarming detail: AI agents harvested more than 2,100 Azure AD token sets from over 40 tenants in about 34 hours — the credential-theft step is already automated. Recorded Future has already tracked more attacks in 2026 than in all of 2025.

For businesses, the conclusion is blunt: there are more adversaries, younger and faster, but the playbook hasn't changed — phishing emails, stolen credentials, and unpatched entry points remain the main ways in. The UK National Cyber Security Centre said something similar in its warning earlier this month: AI ransomware could hit critical infrastructure within two years. The third-quarter data says the fight is already here, and defense still comes down to the basics: multi-factor authentication, credential rotation, backups, and drills.

Recommended Tools

More