Researchers used Anthropic’s Claude to hack into OpenAI
Article image or reusable cover for TechCrunch AI
Security researchers at startup Hacktron AI used Anthropic's Claude to break into OpenAI's systems as part of a bug-bounty program.
They chained together two critical vulnerabilities—first exploiting an image-upload flaw in the Discourse forum software, then leveraging a memory bug in the libheif library that handles Apple image formats. After gaining access to OpenAI employee ChatGPT accounts, they reached the company's code repository. OpenAI paid Hacktron $6,500 for reporting the vulnerabilities, and the company has since fixed both security issues.
Vibekollen prepared this summary with AI from the original publication. The content belongs to TechCrunch AI.
More from TechCrunch AI
Google froze its open source bug bounty program due to a ‘significant rise’ in AI submissions
TechCrunch AI 15 h ago
Can ‘super intelligence’ and a non-binding safety pact solve AI’s image problem?
TechCrunch AI 15 h ago
Trump unveils his new Super Intelligence Force
TechCrunch AI 20 h ago
Amazon responds to data center backlash, says it no longer uses NDAs
TechCrunch AI 3 Oct