AI brief
A team of three independent security researchers at Hacktron used Anthropic's Claude Opus 4.8 and 5 to access OpenAI employee accounts and OpenAI's GitHub repos in under 72 hours, according to The Wall Street Journal.
Why it matters: It shows AI models being used as a tool in attacks against a major AI company's internal systems.
Written by AI from The Verge AI's published text. Read the original for full details.
More coverage of this story (4)
Researchers used Anthropic's Claude to hack into OpenAI
Researchers used Anthropic’s Claude to hack into OpenAI
TechCrunch AI · 18 Sept 2026, 14:00
Researchers used Claude to hack OpenAI
Ars Technica AI · 18 Sept 2026, 13:30
OpenAI ‘ethically hacked’ with help of Anthropic’s Claude chatbot
Guardian Technology · 18 Sept 2026, 11:20
White hat hackers just breached OpenAI using Anthropic's Claude in less than 72 hours — and it is a case study in just how fast AI is advancing
TechRadar AI · 18 Sept 2026, 11:15