techcrunch.com web signal

Researchers used Anthropic's Claude to hack into OpenAI | TechCrunch

By Alexis Dufresne Published September 18, 2026 at 18:12 UTC

Summary

Researchers used Anthropic's Claude to hack into OpenAI | TechCrunch

Shared on Bluesky by 2 AI experts

Scott McGrath @smcgrath.phd: Anthropic's Claude Opus 5 was used by researchers to breach OpenAI, chaining image-conversion flaws to access employee accounts. An earlie… →
Stefanie Hane @stefihane.bsky.social amplified

@techcrunch.com

Researchers used Anthropic’s Claude to hack into OpenAI
View on Bluesky →

Originally reported by techcrunch.com

Read the original article →

Original headline: Researchers used Anthropic's Claude to hack into OpenAI | TechCrunch