A hacker has leveraged artificial intelligence to orchestrate an extensive cybercrime operation, according to a report released by Anthropic on Tuesday (August 27). The hacker used Anthropic’s Claude AI chatbot to carry out what is described as the most comprehensive AI-driven cybercriminal activity to date, targeting at least 17 organizations, including those in healthcare, emergency services, and government sectors.
The report highlights that the hacker used Claude Code to automate various stages of the attack, from reconnaissance to harvesting credentials and penetrating networks. The AI was also used to make strategic decisions, such as determining which data to exfiltrate and crafting targeted extortion demands. Rather than encrypting stolen data with traditional ransomware, the hacker threatened to publicly expose the information, demanding ransoms that sometimes exceeded $500,000.
Anthropic’s report underscores the evolving nature of AI-assisted cybercrime, noting that such tools can now provide both technical advice and operational support, reducing the need for extensive technical expertise. This development poses significant challenges for defense and enforcement, as AI tools can adapt to defensive measures in real time.
In response to the attack, Anthropic banned the accounts involved and implemented new detection methods to prevent future incidents. The company has also shared technical indicators of the attack with relevant authorities to help mitigate similar threats elsewhere.
Recent Comments