At a pivotal moment in technology, AI is no longer a distant promise—it’s actively reshaping the cyber threat landscape. Jeetu Patel, President and Chief Product Officer, Cisco, paints a vivid picture of the challenges and opportunities enterprises face as AI systems grow increasingly autonomous.

The Agentic Challenge

“Everything compounds exponentially with agentic systems,” Patel explains. With thousands of AI agents communicating and acting autonomously, traditional approaches to identity, access, and authentication are no longer sufficient. “If you say I’m going to use my voice as my ID, that’s very easily replicable now,” he notes. “We need a fundamentally different architecture to secure agentic systems—and even traditional systems—from these new threats.”

Patel outlines Cisco’s layered response: Predictive attack detection, identity stores that cover humans and machines, and a zero trust architecture extended to agents. Security is no longer a perimeter concept but “hyper-distributed, sprinkled across the network.” Each packet, from endpoint to host, can be observed for anomalies, making proactive defense possible.

He emphasizes that cybersecurity cannot lag behind the sophistication of threat actors. The industry currently struggles with AI adoption due to high costs and low efficacy of generic models. Cisco’s response: Foundation-sec, a purpose-built, open-source security model optimized for efficiency and predictive accuracy. “You can now run it on a laptop CPU, with performance rivaling much larger models,” Patel says, highlighting the balance between cost and effectiveness.

Data and Network Convergence: A Competitive Advantage“Security is a data game,” Patel asserts. By correlating data across email, web, endpoints, and networks, Cisco can detect and remediate threats before they escalate. The combination of Cisco’s networking and security expertise, now enhanced by the Splunk acquisition, allows for end-to-end SOC reimagination—a rare, durable competitive advantage in a fragmented market.

For Patel, identity intelligence is paramount. Most breaches today happen through stolen credentials, not break-ins. Securing AI models themselves is equally critical, given their unpredictable behavior. “You have to enhance visibility, validate models, and enforce runtime guardrails,” he says. These measures ensure consistent security across applications, agents, and AI models.

CISOs globally face significant challenges: complexity, talent gaps, infrastructure limitations, and rapid AI evolution. Patel stresses that trust becomes a prerequisite for AI adoption. Nations and enterprises must navigate regulations, data sovereignty, and public-private partnerships carefully to maintain security and economic competitiveness.

Personal Playbook: Learning at the Speed of Change

At a personal level, Patel emphasizes relentless learning. “I learn more today than I ever did in my twenties,” he says. His nightly three-hour learning sessions and the principle of improving 1.27% every day illustrate the power of compounding knowledge—a mindset he encourages across Cisco’s teams.

Looking Ahead

Closing the conversation, Patel underscores Cisco’s evolving role. “This is a very new Cisco,” he says. “We are the critical AI infrastructure company for the world,” uniquely positioned to fuse networking, security, and AI into solutions that protect enterprises today and anticipate threats tomorrow.

Note: The journey of AI adoption, as seen in this story, is exactly what the Making AI Work Summit & Awards represents. This leading AI conference fosters dialogue, collaboration, and innovation, while its integral recognition platform—the ET AI Awards India—celebrates the individuals and organisations showcased at the Summit for their groundbreaking contributions to AI. One platform, one purpose: to make AI work for business and society.

Join the community of 2M+ industry professionals.

Subscribe to Newsletter to get latest insights & analysis in your inbox.

All about ETCISO industry right on your smartphone!