• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
Technology

Researchers reportedly use Claude to breach OpenAI's internal system

Forbes reports the breach was part of a bug bounty program. Hacktron's founder, quoted by MTSlive, says Claude Code needed human expertise and guidance—not just an instruction to hack OpenAI.

ForbesFO
Sad CreatorSC
NEWSMAXNE
4 Sources, 21d ago, first seen 21d ago

TLDR

Forbes reports that cybersecurity researchers used Anthropic's Claude to breach OpenAI's internal system through a bug bounty program. A researcher disclosing “HEIF Heist” says the months-long investigation into libheif, an image library used by many apps, enabled hacks of OpenAI and other platforms. MTSlive quotes Hacktron's founder saying Claude Code did not hack OpenAI on its own: prior knowledge of OpenAI's infrastructure helped direct the work, and the AI still needed human guidance.

Combined views

67.2K

4 Sources, first seen 21d ago

419 likes35 comments75 saves90 reposts

Combined views

67.2K

4 Sources, first seen 21d ago

419 likes35 comments75 saves90 reposts

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

4 Sources

Forbes@ForbesA group of cybersecurity researchers managed to breach OpenAI’s internal system using rival Anthropic’s Claude AI software as part of a bug bounty program, highlighting the growing threats posed by advanced AI models. Read more: https://www.forbes.com/sites/siladityaray/2026/09/18/security-researchers-hacked-into-openai-using-anthropics-claude/?utm_source=ForbesMainTwitter&utm_medium=social&utm_campaign=ForbesMainTwitter 📸: Matteo Della Torre/NurPhoto via Getty Images21d
Sad Creator@SadCreatorTalksHacktron AI used Anthropic's Claude to ethically hack OpenAI. Started on the staff Discourse forum, chained a HEIF/libheif bug into SSO, got into employee ChatGPT/Codex accounts, then opened a harmless PR on OpenAI's private GitHub as proof. OpenAI paid $6,500. Opus 4.8 couldn't write the working exploit. Overnight Opus 5 could. Still deciding if that's funny or terrifying.21d
NEWSMAX@NEWSMAXIndependent security researchers reportedly used Anthropic's Claude technology to hack OpenAI's ChatGPT systems, the latest sign that the complexity of computer systems makes them difficult to defend. https://www.newsmax.com/us/ai-artificial-intelligence-hacktron-ai/2026/09/18/id/1269890/21d
MTS@MTSliveHacktron founder @S1r1u5_ says Claude Code didn’t hack OpenAI on its own, it took a human who had already found a Discord zero-day to know where to look: "Everyone has the same Claude Code. It doesn't mean that they'll be doing this. My mom can have Claude Code, that doesn't mean she will start hacking now." "@rootxharsh has a pretty good understanding of OpenAI infrastructure, and he previously found a zero day on Discord, the bug we exploited in the third party. That kind of intuition and previous understanding pointed him at the target." "It's not like go hack OpenAI would make Astra find this kind of vulnerability. Model capabilities are not there. You still need to organize them, hand hold them, point at the right things." "Probably GPT-8 doesn't need that. But GPT-6 or GPT 5.6, they're not there. You can't just give a long-running task which runs for six months where your goal is to hack OpenAI." @HacktronAI21d
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    4 Sources

    Forbes@ForbesA group of cybersecurity researchers managed to breach OpenAI’s internal system using rival Anthropic’s Claude AI software as part of a bug bounty program, highlighting the growing threats posed by advanced AI models. Read more: https://www.forbes.com/sites/siladityaray/2026/09/18/security-researchers-hacked-into-openai-using-anthropics-claude/?utm_source=ForbesMainTwitter&utm_medium=social&utm_campaign=ForbesMainTwitter 📸: Matteo Della Torre/NurPhoto via Getty Images21d
    Sad Creator@SadCreatorTalksHacktron AI used Anthropic's Claude to ethically hack OpenAI. Started on the staff Discourse forum, chained a HEIF/libheif bug into SSO, got into employee ChatGPT/Codex accounts, then opened a harmless PR on OpenAI's private GitHub as proof. OpenAI paid $6,500. Opus 4.8 couldn't write the working exploit. Overnight Opus 5 could. Still deciding if that's funny or terrifying.21d
    NEWSMAX@NEWSMAXIndependent security researchers reportedly used Anthropic's Claude technology to hack OpenAI's ChatGPT systems, the latest sign that the complexity of computer systems makes them difficult to defend. https://www.newsmax.com/us/ai-artificial-intelligence-hacktron-ai/2026/09/18/id/1269890/21d
    MTS@MTSliveHacktron founder @S1r1u5_ says Claude Code didn’t hack OpenAI on its own, it took a human who had already found a Discord zero-day to know where to look: "Everyone has the same Claude Code. It doesn't mean that they'll be doing this. My mom can have Claude Code, that doesn't mean she will start hacking now." "@rootxharsh has a pretty good understanding of OpenAI infrastructure, and he previously found a zero day on Discord, the bug we exploited in the third party. That kind of intuition and previous understanding pointed him at the target." "It's not like go hack OpenAI would make Astra find this kind of vulnerability. Model capabilities are not there. You still need to organize them, hand hold them, point at the right things." "Probably GPT-8 doesn't need that. But GPT-6 or GPT 5.6, they're not there. You can't just give a long-running task which runs for six months where your goal is to hack OpenAI." @HacktronAI21d
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet