- Anthropic claims to have discovered an entirely AI-based cyberattack targeting at least 30 organizations
- This would be the first attack without substantial human intervention.
- Threat actors are increasingly using AI in their attacks, as are defenders.
Anthropic researchers discovered that its own model, Claude, was being used, as an agent, not only as an advisor, but also to execute a cyberattack himself.
The company claims the highly sophisticated cyberespionage campaign manipulated Anthropic’s Claude Code tool to attempt to infiltrate approximately 30 global targets, primarily targeting large technology companies, government agencies and financial institutions.
Although AI-generated cybersecurity threats hitting businesses are now a familiar story, what makes this attack different, in Anthropic’s own words, is: “We believe this is the first documented case of a large-scale cyberattack executed without substantial human intervention.”
China-related activity
Anthropic says it has now assessed with “high confidence” that this threat actor was a Chinese state-sponsored malicious actor. Once the activity was detected, an investigation determined that the attack relied on several AI model features that were not available in 2024.
The intelligence required to follow complex commands and coding tasks in context is a feature that has only recently reached the required level – and it has been coupled with recently developed tools that allow them to crack passwords, analyze networks, and more.
Perhaps most important is the agency that models now have. The rise of agentic AI means that AI can accomplish these tasks on its own with only minimal, occasional input: making decisions and operating in a loop.
“Agents are valuable for daily work and productivity, but in the wrong hands they can significantly increase the viability of large-scale cyberattacks,” explains Anthropic.
“This campaign has substantial implications for cybersecurity in the era of AI “agents,” systems that can operate autonomously for extended periods of time and that perform complex tasks largely independent of human intervention.
Security teams and hackers are well aware of the transformative power of AI as the cyber landscape has been reshaped by automated threat deployment and detection tools.
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds. Make sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp Also.
The best identity theft protection for every budget
