GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents

Summary

OpenAI introduced GPT-6 Cyber, a cybersecurity model that scores 48.5% on ExploitGym, up from 32.1% for GPT-5.6 Cyber, and avoided exploiting out-of-scope services in 99.93% of tests without its deployment prompt and 100% with it. Its researchers used it to develop 948 Linux privilege escalation exploits and identify 81 previously unknown high-severity Firefox vulnerabilities, and Mozilla has released fixes for many of them.

Mitigation

Mozilla has released fixes for many of the Firefox vulnerabilities; the model also helped develop candidate patches. No other mitigation is specified in the source text.

Related items

• InfoPerplexity launches its own take on AI agents for workSame vendor · The Verge (AI)
• MediumBinance heist is stark example of AI governance gaps and basic security failuresSame vendor · CSO Online
• MediumFake OpenAI, SpaceX events used to steal Microsoft, Google accountsSame vendor · BleepingComputer
• LowOpenAI’s GPT-6 Astra silently obstructed safety research tasksSame vendor · BleepingComputer
• InfoWinner of the Prix Goncourt literary prize admits to using AISame vendor · The Guardian Technology