AI Governance: Beyond Browser to Network-Level Controls

Organizations are struggling to control sensitive data as employees use artificial intelligence tools. Paul Martini, co-founder and CEO at iboss, highlights that employees can expose company intellectual property by using personal AI accounts. Autonomous agents also introduce risks by accessing corporate data. The issue is that these agents work with company data. Traditional AI governance often focuses only on browser activity, missing interactions within applications. Martini says organizations need network-level controls. These controls can inspect data moving between devices and AI services, identify sensitive information, and prevent unauthorized transfers. What's interesting is that AI governance must extend beyond browser-based tools to include applications with built-in AI capabilities. AI can also improve data loss prevention by reducing false positives and extracting richer signals from data flows. The rapid spread of new applications is challenging current security approaches. This matters because protecting sensitive company data from AI-related risks is becoming increasingly complex.