At the RSA Conference 2026, Cisco’s Jeetu Patel revealed that while 85% of enterprises are piloting AI agents, only 5% have deployed them into production, primarily due to a lack of trust in these systems. This trust deficit, which Patel likened to the immaturity of teenagers, highlights the industry’s shift from managing information risks, such as those associated with chatbots, to facing action risks where AI agents can take irreversible actions. Patel emphasized that establishing a robust trust architecture is critical for enterprises as they navigate this transition, and noted Cisco’s recent integration of its Defense Claw framework with Nvidia’s OpenShell secure container in under two days as a step toward addressing these challenges.

Cisco: Cisco provides enterprise networking, security, and collaboration solutions with a growing emphasis on AI-driven infrastructure. At RSA Conference 2026, it addressed the enterprise AI agent trust deficit through launches like Defense Claw and AI Defense Explorer Edition, enabling secure agent deployment. The company mandated AI-built products across its engineering organization to maintain competitive speed.
Jeetu Patel: Jeetu Patel is Cisco’s President and Chief Product Officer, directing global product strategy and innovation. In a VentureBeat interview at RSA Conference 2026, he explained the gap between AI agent pilots and production as a matter of trusted delegation, comparing agents to immature teenagers needing guardrails. He highlighted Cisco’s rapid open-source integrations and development mandates as paths to market leadership.
Defense Claw: Defense Claw is Cisco’s open-source security framework that bundles tools for scanning and protecting AI agent deployments. Developed and integrated with Nvidia’s OpenShell within days of its announcement ahead of RSAC 2026, it automates security enforcement at container launch. This positions it as a key enabler for trusted AI agent operations in enterprises.
AI Defense Explorer Edition: AI Defense Explorer Edition is Cisco’s free self-service red teaming tool for testing AI agent vulnerabilities. Unveiled at RSAC 2026, it allows enterprises to proactively assess agent workflows against threats before production. It supports Cisco’s strategy to bridge the trust gap in scaling AI agents.

`json
{
“Security Shift”: “The industry is moving from concerns over information risks with chatbots to action risks with autonomous AI agents taking irreversible actions.”,
“Cisco Innovation”: “Cisco showcased its ability to quickly integrate Defense Claw with Nvidia’s OpenShell secure container within a notably short time frame.”,
“Industry Challenge”: “A significant issue for enterprises is the ongoing gap between piloting AI agents and moving them to production, primarily due to the lack of trust architectures.”
}
`