OpenAI reportedly preparing to unveil new GPT-6 Cyber model
OpenAI is reportedly preparing to preview a new artificial intelligence model designed specifically for cybersecurity, alongside a new product aimed at helping customers deploy it securely.
GPT-6 Cyber could be unveiled within days, potentially at OpenAI’s annual DevDay event in San Francisco on 29 September, according to Fortune.
Listen to this article
The model is designed to help organisations respond to increasingly sophisticated cyber threats, including by identifying and patching vulnerabilities and supporting automated security workflows.
A limited number of customers are reportedly already testing GPT-6 Cyber through OpenAI’s application-only Daybreak Red programme, which provides vetted partners with access to its most advanced cybersecurity tools.
Alongside the model, OpenAI is reportedly developing a yet-to-be-named product intended to make it easier for customers to deploy GPT-6 Cyber in a secure and automated way, while giving the company greater oversight of how the technology is being used.
GPT-6 Cyber would be OpenAI’s fourth cybersecurity-focused model this year, following GPT-5.4 Cyber in April, GPT-5.5 Cyber in June and GPT-5.6 Cyber in August.
The reported launch comes as concerns grow over the potential use of increasingly capable AI systems in cyberattacks.
At the same time, OpenAI has been expanding its focus on defensive cybersecurity tools and recently announced plans to invest $1 billion to subsidise access to its products for critical services.
OpenAI has not officially commented on GPT-6 Cyber, but Fortune reports that DevDay could see the company announce a dozen or more new products across its consumer and enterprise businesses.