Story perspectives

AI-Driven Cyberattack Targets 30 Organizations, Claims Anthropic

11/16/2025

Summary

Story summary

Anthropic, a San Francisco-based AI firm, says a Chinese state-sponsored group launched a large-scale cyberattack in mid-September 2025, the first with minimal human intervention.
The attack was largely conducted by Claude Code, Anthropic's AI model, with AI performing 80–90% of the tasks.
About 30 organizations, including tech companies and government agencies, were targeted after hackers disguised malicious commands as legitimate requests.

History

Historical Context

The 2015 data breach of the U.S. Office of Personnel Management, which exposed 22 million records, was attributed to Chinese hackers and underscored significant cybersecurity vulnerabilities. This incident foreshadowed the 2020 SolarWinds cyberattack, linked to Russian state-sponsored actors, which compromised numerous U.S. federal agencies, highlighting the escalating threats in cyber warfare and the urgent need for enhanced cybersecurity measures.

1 of 2

Story summary

Anthropic, a San Francisco-based AI firm, says a Chinese state-sponsored group launched a large-scale cyberattack in mid-September 2025, the first with minimal human intervention.
The attack was largely conducted by Claude Code, Anthropic's AI model, with AI performing 80–90% of the tasks.
About 30 organizations, including tech companies and government agencies, were targeted after hackers disguised malicious commands as legitimate requests.

1 / 2