Media Coverage

The image of the bank robber is hopelessly outdated. Today’s heist does not begin with a getaway car outside a branch. It begins quietly, with an adversary establishing persistence inside a financial institution’s network and studying how the organisation responds, says Tom Kellermann, VP of AI Security and Threat Research at Trend AI.

The cumulative milestone reached by TrendAI, whose U.S. headquarters is in Irving, underscores how quickly AI is reshaping cybersecurity spending — and how the AWS Marketplace captures that growth.

Tom Kellermann, VP of Al Security and Threat Research, analyses how Al has altered the cyberattack kill chain, as enterprise Al is being used against them

Security teams often treat frontier and open-weight artificial intelligence models as competing choices. The two are complementary, not substitutes, said Johnny Hand, global CISO at TrendAI.

The number of software security flaws discovered in popular technology products in 2026 is on pace to roughly double the tally of vulnerabilities that surfaced in 2025, an explosion driven by increasingly capable artificial intelligence systems.

A jailbroken Google Gemini did 90 percent of the work in a credential- and cryptocurrency-stealing spree, including spinning up a new command-and-control (C2) server in just six minutes, according to a TrendAI report shared exclusively with The Register.

TrendAI has published research on a cybercrime economy built around stolen healthcare data, ransomware, and traded access to health systems.

Economic growth and innovation thrive in the light. But mounting cyber risk threatens to cast a shadow over global organizations at a time of tremendous change. From healthcare to financial services, and energy to education, AI is helping businesses to work faster, smarter and more efficiently. But it also exposes those same organizations to risk.

New research from TrendAI highlights the immense reach of Fancy Bear, also known as APT28 and Forest Blizzard.

TrendAI finds 70% of requested exploits in the last three years were for vulnerabilities that were at least two years old.