AI-Built Exploit and Sign-In Flaw Opened Path to Internal OpenAI Code

Researchers at security firm Hacktron used Claude to build a working exploit for a vulnerability in an image-processing library, then chained it with a flaw in OpenAI’s sign-in system to take over employee ChatGPT and Codex accounts, and ultimately gained access to internal code repositories.

The entry point was OpenAI’s community forum, community.openai.com, which runs on Discourse. Discourse’s built-in image checks didn’t support the HEIC/HEIF photo format, so uploads in that format were passed to ImageMagick, exposing an unpatched flaw in the libheif library it relies on for decoding. 

Hacktron says the underlying bug had been fixed upstream a year earlier without ever being flagged as a security issue, so it was never assigned a CVE and missed the usual patching cycle.

Turning the flaw into a reliable exploit took several attempts and involved Claude Opus 4.8 and Opus 5. The exploit allowed remote code execution, which the researchers first used against a test Discourse instance, then on OpenAI’s own forum.

Because the forum let people sign in with their OpenAI account, code execution there opened a path to broader account access. Hacktron says that until the issue was fixed, any user or employee who logged into the forum could have had their ChatGPT and Codex accounts taken over.

In addition, since people often connect other services to those accounts, the theoretical exposure extended to services like GitHub, Slack and email.

OpenAI distinguishes between the two flaws. It told SecurityWeek that the image-processing bug lived in the third-party service Discourse, while the account-takeover path was a separate, OpenAI-side issue. 

The OpenAI weakness was related to sign-in tokens generated for the community forum carrying excessive permissions and granting full API access to associated ChatGPT and Codex accounts.

To demonstrate the access without reading any internal code, Hacktron says it took over an OpenAI employee’s account whose Codex integration was linked to OpenAI’s GitHub organization, then used it to open a pull request in an internal repository before stopping further testing. 

OpenAI said its own review of the incident found limited reads of private-repository metadata and commits, followed by the researcher-submitted pull request, specifically to a README file.

Hacktron’s report also raises Slack as a service that could theoretically have been reached through connected accounts. OpenAI says Hacktron did not verify actual access to employee Slack messages.

The security firm reported the account-takeover issue to OpenAI through Bugcrowd, and the AI giant confirmed a fix about 14 hours later. It separately reported the libheif flaw to Discourse through HackerOne. Discourse had a fix ready within two days and added image-processing sandboxing as an extra layer of defense, then published a security advisory.

In a statement, OpenAI said, “We thank the researchers for contacting us and sharing their findings. We narrowed the permissions on Community sign-in tokens and revoked affected tokens and sessions.” 

The company paid Hacktron a $6,500 bounty for the OpenAI-side finding. 

Related: OpenAI Says Its Models Searched GitHub for Leaked API Keys During Training

Related: AI Agents Can Retrain Own Models Mid-Task, Leaking Secrets and Erasing Refusals

Related: OpenAI Investigates Report Linking AI Agents to RubyGems Attack