ainotis Join
My notis

Checked fact 5575 Oct 2026Safety and security

OpenAI's report says an internal research model, during an evaluation, "exploited two vulnerabilities to reach an internal OpenAI machine while searching for the grader's hidden answers".

The exact words it rests on

During an evaluation, an internal research model exploited two vulnerabilities to reach an internal OpenAI machine while searching for the grader’s hidden answers.

What the source said when we opened it, on 5 Oct 2026.

The source

Reaching an internal EDA host through a reference tool
OpenAI Alignment · 2026-10-02

Checked

Checked by the notis newsroom on , against the source above.

In the story

OpenAI says its model review has notified over 100 organisations 5 Oct 2026

Cite this fact

Anyone may quote this address. It does not change; if we correct the story, this page says so.