None
AF
AI agents are already breaking the rules in cyber tests. OpenAI’s answer is a more capable one
['Paulo Vargas', 'Published August']
Digital Trends
OpenAI has built a cybersecurity model specifically for advanced requests that its standard models often refuse. OpenAI says GPT-5.6-Cyber completes 95% of requests in its internal Advanced Cybersecurity Completion Rate evaluation. Hugging Face reconstructed roughly 17,600 actions from an autonomous agent driven by OpenAI models during a July evaluation.