ChatGPT and Codex flaws patched Feb 2026 exposed DNS exfiltration and GitHub tokens, raising enterprise AI security risks.
What happens when researchers think outside the box? Data gets exfiltrated through DNS.
Command injection in Codex and a hidden outbound channel in ChatGPT exposed risks of credential theft and covert data ...
9don MSN
Security experts discover critical flaw in OpenAI's Codex able to compromise entire organizations
Researchers managed to steal GitHub OAuth tokens by abusing a command injection vulnerability.
OpenAI patches ChatGPT vulnerability that allowed silent data exfiltration and fixes Codex flaw exposing GitHub tokens.
OpenAI has introduced GPT-5 Codex, a version of GPT-5 that is fine-tuned for coding agent workflows and built to better support developers working across varied environments. Rather than just ...
The plugin allows developers to run Codex reviews and delegate tasks directly within Anthropic’s Claude Code environment ...
Codex struggles with big-picture debugging in complex codebases. Deep Research excels at diagnosis when code context is large. Human testing and oversight still remain critical with AI coding. "Huh?!?
OpenAI published a Codex plugin on March 30 that installs directly inside Anthropic’s Claude Code, letting developers run code reviews and delegate tasks to Codex without leaving their existing ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results