AI Fuels Surge in Cybercrime and Ransomware Threats

Artificial intelligence is increasingly being weaponised by cybercriminals. Generative AI helps attackers locate software vulnerabilities, write bespoke malware and launch highly personalised phishing and voice‑spoofing attacks, cutting breach preparation time. Verizon’s 2026 Data Breach Investigations Report notes software flaws now initiate almost a third of breaches, overtaking stolen passwords. Ransomware activity rose 48% in May 2026, with the education sector hit hardest, followed by government and telecoms. Notable incidents include a 7‑Eleven franchise‑document breach and an extortion attack on Foxconn that exposed over 11 million files.

Experts say AI removes skill barriers, turning cyber conflict into an autonomous system where humans set strategy but machines execute at scale. They urge a shift from static defenses to continuous governance of AI models to mitigate new attack surfaces such as prompt injection and data poisoning. In parallel, U.S. officials drafted the Great American AI Act and barred foreign nationals from Anthropic’s latest tools over security concerns, while GitHub disclosed a breach of internal code repositories.