Edition November 6, 2026

OpenAI’s GPT-6 Cyber can break into systems and propose fixes

OpenAI says GPT-6 Cyber is designed to find software security weaknesses, create working exploits and recommend patches.

In a targeted campaign, GPT-6 Cyber found 81 previously unknown severe vulnerabilities in Firefox, even though earlier AI models had already tested the browser, OpenAI says. The model also found that 30 known issues were more severe than previously believed; Mozilla accepted the reports and has shipped fixes. The benchmark figures come from OpenAI alone; no independent testing of the model has been published so far.

This summary is based on the available full source text. Summary prepared with AI assistance. The original source is authoritative.