Wire · technology
OpenAI unveils GPT-6 Cyber: Astra capabilities for security work
◆ Sectors
◆ Geography
◆ Source
◆ Verified
Fusion42 · 6 November 2026 · Fusion42 review
OpenAI has launched GPT-6 Cyber, an advanced AI model designed to identify and validate cybersecurity vulnerabilities with higher accuracy and better safety controls than its predecessor, GPT-5.6 Cyber. The model has discovered hundreds of new exploits in Linux and Firefox, is restricted to trusted organisations via the Daybreak initiative, and competes with similar AI security tools from Google and Anthropic.
This Wire brief sits within Fusion42's coverage of Cybersecurity and AI Frontier Models, and 13 sources have reported it between 5 Nov 2026 and 17 Nov 2026.
◆ ◆ The Wire takeaway
You face a new security AI that finds exploits faster and safer than before. Licenced access narrows your competition but raises the bar for your defensive tech this week.
◆ Coverage
13 sources · first reported 5 Nov 2026 · latest 17 Nov 2026
◆ Related on Wire
◆ Topics