White Hats Used Anthropic’s Claude to Break Into OpenAI in 72 Hours
A small team of researchers chained three bugs with Anthropic’s Claude Opus 5 and ended up inside OpenAI’s private code repository. The whole job took under 72 hours, and OpenAI paid $6,500 for it. The Chain, Link by Link A vulnerable ‘libheif’ library inside Discourse, the forum software, allowed remote code execution (RCE) on the […]
Key Takeaways
- A small team of researchers chained three bugs with Anthropic’s Claude Opus 5 and ended up inside OpenAI’s private code repository.
- The whole job took under 72 hours, and OpenAI paid $6,500 for it.
- The Chain, Link by Link A vulnerable ‘libheif’ library inside Discourse, the forum software, allowed remote code execution (RCE) on the […]
Summary generated by IndiCrypto from News.Bitcoin.com. IndiCrypto is a news aggregator and does not provide investment advice. Read the original article for full context.
More in Security
Blockchain malware activity jumps 440% as AI lowers the barrier for North Korea and Iran-linked hackers
Public chains preserve malware instructions beyond ordinary domain takedowns, shifting defense toward monitoring wallets, contracts and off-chain servers. The post Blockchain malware activity jumps 440% as AI lowers the barrier for North Korea and Iran-linked…
Nostra Joins a September That Has Already Cost Crypto Over $326 Million
Starknet lending protocol Nostra paused its money market on Thursday after a manipulated price oracle let one account borrow roughly $3.5 million against NSTR collateral. Lending, borrowing, withdrawals, and liquidations remain unavailable while the team…
OpenAI Models Are Writing Their Own Jailbreak Instructions—And Sometimes Obeying Them
OpenAI's new transparency framework reveals AI models that invented fake "breach alerts," coached themselves to hide mistakes, and smuggled a file onto the public internet to talk to each other.