---
title: "OpenAI used AI agents to hack its own defenses before resuming model training"
date: 2026-10-24
source: https://www.calcalistech.com/ctechnews/article/mi58wfzh3
description: "The company spent hundreds of billions of tokens testing whether teams of AI agents could escape sandboxes, reach the live internet or communicate across isolated environments. The exercise uncovered software vulnerabilities and configuration errors that OpenAI says it has addressed."
---

# OpenAI used AI agents to hack its own defenses before resuming model training

The company spent hundreds of billions of tokens testing whether teams of AI agents could escape sandboxes, reach the live internet or communicate across isolated environments. The exercise uncovered software vulnerabilities and configuration errors that OpenAI says it has addressed.

*Published: 2026-10-24*