Anthropic Disrupts First Known AI-Orchestrated Cyber Espionage Campaign Linked to China
- Category: AI Automations
- Date 15-11-2025
AI-powered cyberattacks are getting stronger.
Anthropic said it disrupted the first reported AI-orchestrated cyber espionage campaign, involving agentic capabilities that go beyond just advising on attacks, to executing them autonomously.
Anthropic claims “with high confidence” that the threat actor was a Chinese state-sponsored group that manipulated Claude Code to attempt to infiltrate thirty global targets. The targets include large tech companies, financial institutions, chemical manufacturers and government agencies. Anthropic tracked the severity of the operations, banned associated accounts and notified authorities.
“We’re continually working on new methods of investigating and detecting large-scale, distributed attacks like this one,” Anthropic said in the report.
The incident is just one case exemplifying that hackers are getting far smarter at using AI. Data from Google’s Threat Intelligence Group published in early November showed a shift in hackers’ AI strategies, leaning towards AI-powered malware and models being abused across the attack lifecycle. While the potential for bad actors to leverage this tech was clear from the start, “the speed at which it materialized is unsettling,” Adam Arellano, Field CTO at Harness.io, told The Deep View.
And larger, coordinated attacks may inspire others, Arellano said. “More and more of the smaller groups and even individuals will start to figure out how to use (LLMs) as well, increasing access to these types of attacks.”
But that doesn’t mean that enterprises are empty-handed, Eric O'Neill, former FBI counterterrorism and counterintelligence operative and founder of The Georgetown Group and Nexasure AI, told The Deep View.
Plenty of organizations are deploying AI for counterintelligence, anomaly detection, rapid incident response and resilience in data integrity, he said. But as hackers become more sophisticated at a rapid clip, speed, adaptability and resilience are “the only winning strategies.”
“The battle has escalated into something out of the 1980s film Tron: AI vs. AI, dueling across digital landscapes for control of the currency of our lives—data,” said O'Neill.