Google unveils Sec-Gemini v1, a new experimental AI model focused on cybersecurity

Google has unveiled Sec-Gemini v1, an experimental AI model designed to push the boundaries of cybersecurity AI. This new model integrates Gemini’s advanced capabilities with near real-time cybersecurity knowledge and tools, enhancing its performance in critical areas such as incident root cause analysis, threat analysis, and understanding vulnerability impacts. Sec-Gemini v1 is currently available freely for research to select organizations, institutions, professionals, and NGOs.

Google claims that Sec-Gemini v1 surpasses other models on significant cybersecurity benchmarks. It integrates Google Threat Intelligence (GTI), OSV, and other essential data sources, enabling it to outperform competitors by at least 11% on the CTI-MCQ threat intelligence benchmark and by 10.5% on the CTI-Root Cause Mapping benchmark.