Security researchers used Claude to help them hack into OpenAI

7.9 relevance

Security researchers used Claude to hack OpenAI; highly relevant and actionable.

AI/ML theverge.com

Answer the board on AI readiness

Arrive with evidence, not opinion. See how your foundation benchmarks against your peers, get a clear read, and leave with a ranked place to start.

Why It Matters

For engineers building AI infrastructure, this demonstrates how LLMs can accelerate sophisticated vulnerability chains, making third-party dependencies like Discourse and image processing libraries critical attack surfaces that demand rigorous auditing and runtime protection.

Key Takeaways

Audit all third-party integrations (forum software, image parsers) for exposure to LLM-assisted exploitation, and implement detections for anomalous pull requests from compromised accounts.

Summary

A three-person Hacktron team used Anthropic's Claude Opus 4.8 and 5 to hack into OpenAI in under 72 hours, exploiting a HEIF image processing flaw in Discourse Cloud to achieve RCE and access the Monorepo containing algorithmic secrets. They sent a pull request from an employee's Codex account to prove access, spent less than $3,000 in tokens on the campaign, and were only detected by Shopify among multiple targets. OpenAI later paid them a $6,500 bug bounty.

Author

Stevie Bonifield