AI Tools & Products
6d ago
OpenAI launches GPT-6 Cyber model for cybersecurity professionals amid rising AI threats
Nov 5, 2026
AI Summary
OpenAI has introduced GPT-6 Cyber, an AI model designed to assist cybersecurity experts, outperforming its general-purpose counterpart in various tasks. Due to potential misuse, access is currently limited to a select group of approved security customers, as businesses face increasing pressure to enhance their cyber defenses amid a surge in AI-driven attacks.
- OpenAI has launched a new AI model called GPT-6 Cyber, aimed at cybersecurity professionals.
- The model is reported to outperform the general-purpose GPT-6 Astra on cybersecurity tasks.
- Access to GPT-6 Cyber is restricted to a small number of approved security customers to mitigate risks of misuse.
- OpenAI's move comes as AI models improve at identifying software vulnerabilities, prompting businesses to strengthen their cybersecurity measures.
- Gartner forecasts global AI cybersecurity spending to reach $51.3 billion this year, nearly double last year's figure.
- This is OpenAI's fourth cybersecurity-focused model released in 2023, following earlier versions in April, June, and August.
- OpenAI has hired Dali Rajic as chief revenue officer, who has a background in cybersecurity software.
- Competitors like Google and Anthropic are also targeting the cybersecurity market with their own models and programs.
- GPT-6 Cyber specializes in advanced vulnerability research and exploit development, allowing researchers to test server defenses.
- OpenAI tested the model on Firefox and the Linux kernel, discovering numerous high-severity vulnerabilities.
- Researchers found 81 new vulnerabilities in Firefox and developed exploits for 948 Linux kernel vulnerabilities, significantly more than previous models.
- OpenAI offers two access levels through its Daybreak program: Daybreak Red for offensive tasks and Daybreak Blue for defensive tasks.
- Initially, GPT-6 Cyber will be available only to select Daybreak Red customers, with plans to expand access in the coming weeks.
cybersecurityai modelcyber attacksopenaigpt-6