Research

LLMs, Ransomware and the AI Boom’s Cybersecurity Challenges

Page published

Publication date unknown · First observed: 18 Nov 2026

Coverage timeline

18 Nov 2026nyu.eduobserved

Single-source research — one report is available.

Why it matters

PromptLock research highlights how LLMs could reduce the human effort required for ransomware by automating attack decisions as well as code generation.

NYU Tandon researchers describe PromptLock, a simulated malicious AI system developed to explore LLM-driven ransomware orchestration, including malicious code generation and attacker decision-making. The researchers distinguish their experimental work from real attacks, saying they had not observed LLM-orchestrated attacks.