Google Launches Gemini 3.5 Flash Cyber for AI Security

In a significant move for cybersecurity, Google has announced the launch of Gemini 3.5 Flash Cyber, a specialized AI model built to excel at finding and patching security vulnerabilities. According to a blog post from Google, this new model is positioned as a "cost-efficient and highly capable alternative" to other, often more expensive, large-scale AI systems currently available on the market.

The Gemini 3.5 Flash Cyber is based on the existing Gemini 3.5 Flash architecture and will initially be accessible to government entities and select trusted partners. Its deployment will be facilitated through CodeMender, Google's dedicated coding agent focused on security. Google highlights that CodeMender can leverage 3.5 Flash Cyber multiple times with remarkable speed and minimal cost. This capability allows AI agents to conduct more extensive code scans, thereby increasing the chances of discovering hidden vulnerabilities.

The development signifies Google's commitment to enhancing AI's role in bolstering digital defenses. By offering a more accessible and efficient AI security solution, Google aims to empower organizations, particularly governments, to proactively address emerging threats and maintain the integrity of their digital infrastructure. The potential for rapid vulnerability detection and remediation could mark a new era in how we approach software security, making systems more resilient against cyberattacks.