‘Rogue agents are a new security blindspot’: Vishal Salvi, Global Head, Cybersecurity Services, Cognizant

The rise of autonomous AI agents is prompting enterprises to rethink how they approach security and trust. Unlike earlier AI systems, these agents can reason, make decisions, access data and applications, and act with limited human intervention. This creates a new security challenge – keeping autonomous systems within defined boundaries. Cognizant is helping organisations address this shift by combining security controls, identity management, runtime monitoring and responsible-AI governance to make agentic systems safer to deploy.