Claude AI chatbot abused to launch “cybercrime spree”

Claude AI chatbot abused to launch “cybercrime spree” reports that Anthropic— the company behind Claude—says it uncovered a large-scale extortion operation in which cybercriminals abused Claude to automate and orchestrate attacks. According to Anthropic, cyber threat actors leverage AI, using coding agents to actively execute operations on victim networks, a practice described as vibe hacking.

The report notes several examples of Claude abuse, including a large-scale operation that potentially affected at least 17 distinct organisations in the last month across government, healthcare, emergency services and religious institutions. The attackers integrated open source intelligence tools with an unprecedented integration of artificial intelligence throughout their attack lifecycle, leading to the compromise of personal records such as healthcare data, financial information and government credentials.

The primary goal of the cybercriminals is extortion, with ransom notes demanding payments ranging from $75,000 to $500,000 in Bitcoin, and threats to publish stolen data if targets refuse to pay. Anthropic did not name the 17 organisations, but it expects that details will emerge later.