Phying News
P非影 中文
Curated security research, vulnerabilities, advisories and tools for practitioners.

Hacking OpenAI

hacktron.ai | blog | #supply-chain | #vulnerability-research | #openai | #heap-overflow | #sso-misconfiguration | #internal-repositories

Summary

Researchers describe chaining a heap overflow and an SSO misconfiguration to compromise OpenAI internal repositories.
Published
Collected

original ↗

Related coverage

research · idnsec.com
Mitigating RCE Impact Like in OpenAI Hack with Defense-in-Depth
Using HacktronAI's OpenAI Discourse hack disclosure, this article covers sandboxing, runtime protection, and AI-assisted defense to shrink the blast radius of RCE across interconnected apps.
blog · hacktron.ai
Here’s How an OpenAI Model Went Rogue and Hacked Hugging Face
Using publicly available data, the author reconstructs how an OpenAI model allegedly went rogue and hacked Hugging Face, since neither company fully disclosed the incident.
blog · blog.trailofbits.com
Introducing Patch the Planet
Patch the Planet, a Trail of Bits-OpenAI initiative, sends expert engineers to fix critical open-source projects alongside frontier models; week one yielded 64 pull requests across 19 projects.
blog · xbow.com
GPT-5.5: Democratizing Cyber Capabilities
XBOW endorses OpenAI's broad release of GPT-5.5 over gated access, arguing that private-club gating fails as vendor-sec did, and staged openness with safeguards serves defenders better.
blog · xbow.com
XBOW Unleashes GPT-5’s Hidden Hacking Power, Doubling Performance
Reports that integrating GPT-5 into XBOW's agent framework more than doubled pentest performance, despite OpenAI's system card rating its standalone cyber capabilities as modest.
blog · hackerone.com
Patch the Planet: HackerOne Joins OpenAI's Daybreak Initiative to Secure Critical Open-Source Software
HackerOne is a launch partner in Patch the Planet, a new initiative pairing AI-assisted research with expert human validation to deliver verified fixes to open-source maintainers at scale.
back