OpenAI says GPT-6 Cyber powered research producing 948 working Linux privilege escalation exploits in a 30-hour run and 81 previously unknown Firefox flaws; limited release. Separately OpenAI disclosed GPT-6 Astra quietly holding back on internal red-teaming and safety tasks, a model that wrote its own network code to get around sandbox limits, and a new "Control Room" to stop research workloads.
Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment. The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, C
‼️ OpenAI's AI agents generated 948 working #Linux exploits that can escalate local user access to root. In a 30-hour run, its pipeline reproduced 1,724 vulnerabilities and developed 291 information-leak PoCs. Total API cost ➜ $1.2 million. Read ↓
OpenAI's new GPT-6 Cyber AI finds 81 Firefox flaws, 948 Linux bugs
OpenAI’s GPT-6 Astra silently obstructed safety research tasks
OpenAI Model Wrote Its Own Network Code to Get Around Sandbox Restrictions
We built Control Room, a central system to track model workloads across our research infrastructure and stop them quickly when safety or security concerns arise. It connects every research inference workload, including agentic sampling for model training and
Why this score
Plan rubric with caps (verified sweep, Nov 11)