OpenAI Successfully Ethically Hacked With Assistance From Anthropic's Claude
· Technology · The Guardian, Gadgets360
Cybersecurity researchers at Hacktron AI successfully breached OpenAI employee accounts using Anthropic's Claude chatbot to navigate internal systems. The startup team compromised multiple ChatGPT accounts via a staff discussion forum on the Discourse platform before submitting a harmless pull request to OpenAI's GitHub repository. The researchers noted that while Claude assisted in generating code and planning the attack, they also utilized OpenAI's own advanced GPT-5.6 Sol model during the process. OpenAI acknowledged the findings and confirmed that the exploited vulnerabilities have been addressed. The researchers received a $6,500 payout through OpenAI's bug bounty program.
Why it matters
This incident demonstrates how artificial intelligence tools can dramatically accelerate the discovery of complex software vulnerabilities, highlighting both security risks and the value of bug bounty programs for AI firms.
Read the original report — The Guardian
Join us on Telegram
Breaking news the moment it lands. At 10,000 members we ship the Android app.