Monday, November 23, 2026

Anthropic Reveals Fourth Case of 'AI Hacking External Systems'

Input
2026-09-10 10:22:45
Updated
2026-09-10 10:22:45
Anthropic. Yonhap News Agency

[Financial News] Anthropic has disclosed the fourth case in which an AI model hacked an external system during testing.
According to foreign media reports on the 9th (local time), Anthropic said the incident, which occurred in January, was not discovered until last month. The company emphasized that this highlights the difficulties AI developers face in identifying and controlling the unexpected behavior of advanced models.
Anthropic added that the case involved an early version of Claude Opus 4.6 and that it had notified everyone affected.
Anthropic previously announced in July that some Claude models had hacked the systems of three companies during cybersecurity testing.
Anthropic said the cases involved Claude Opus 4.7, Claude Mythos 5, and another model under internal research. It characterized them as stemming from "a mistake" that inadvertently allowed the AI models to access the internet.
After it became known that an OpenAI AI agent had hacked Hugging Face, an open-source AI-sharing platform, Anthropic reviewed approximately 141,000 tests and identified these three cases in the process.
Anthropic explained that this fourth case was discovered last month in some tests that had been omitted from the review because they were initially deemed unnecessary to examine among the approximately 141,000 tests.
[email protected] Lee Seok-woo, International Affairs Specialist Reporter