India, March 12 -- As agentic AI moves deeper into enterprise security operations, autonomy is no longer theoretical. AI agents are already triaging alerts, prioritising incidents, and executing remediation steps at machine speed. While this promises faster response times and operational efficiency, it also introduces new forms of risk that traditional security models were never designed to handle.

Saugat Sindhu, Global Head of Advisory Services for Cybersecurity and Risk Services at Wipro Limited, says the challenge for enterprises is not whether to adopt autonomous security agents, but how to govern them safely. Drawing on insights from Wipro's State of Cybersecurity Report 2025 (SOCR 2025) and its global client experience, Sindhu outl...