Jun 02 2026
Artificial Intelligence

As Enterprises Deploy More AI Agents, Trust Depends on Visibility

Splunk’s Kamal Hathi says visibility is essential as enterprises deploy autonomous artificial intelligence agents.

Imagine a customer service artificial intelligence agent that has been authorized to issue a $600 refund.

Instead, it refunds $1,500.

The agent wasn’t trying to harm the retailer. It simply made a mistake in a customer service transaction.

That scenario illustrates one of the biggest challenges facing enterprises as they begin deploying autonomous AI agents, says Kamal Hathi, senior vice president and general manager of Splunk.

“Those agents also are running on large language models. They’re running on AI, and sometimes they can make a mistake,” Hathi says.

Click the banner below to apply insights from Cisco Live to your AI strategy.

 

As organizations move beyond AI experimentation and begin allowing agents to take actions on behalf of employees and customers, visibility into those systems is becoming a business necessity. Companies need to understand what AI agents are doing, if they are behaving as intended and if they are operating outside established guardrails.

That need is driving a new focus on observability for AI, Hathi says.

“If you want agentic AI to be practically useful in an enterprise or in an organization, you need to make sure that it scales, that it’s able to deliver results and that you can trust it,” Hathi tells BizTech.

Observability Becomes the Foundation for Trust

Enterprises face three major challenges as agentic AI adoption accelerates, Hathi says: operating at machine scale, operating at machine speed and building trust in AI systems.

The first challenge is managing the explosion of telemetry and machine data generated by AI agents, applications and infrastructure. Organizations need ways to collect and analyze that information while controlling costs and making the data useful for AI-driven operations.

READ MORE: Here are three ways AI is revolutionizing customer experience.

The second challenge is responding to events at machine speed. As AI-generated code, automated workflows and AI-powered threats become more common, security and operations teams need tools that can keep pace with rapidly changing environments.

The third challenge is trust.

“The same unified visibility and telemetry allows us to start looking and seeing what the AI is actually doing,” Hathi says. “Do we understand what it’s supposed to do versus what it’s actually doing?”

For Hathi, observability is the foundation that makes autonomous AI possible. Without visibility into agent behavior, organizations cannot confidently determine whether AI systems are delivering value, operating safely or staying within established business rules.

AI Autonomy Requires Guardrails

The answer is not to limit AI agents, Hathi says, but to create the guardrails needed to manage them.

Using the customer refund example, organizations must establish clear expectations for what an agent is supposed to do and continuously evaluate whether it is behaving accordingly. Those evaluations then become the basis for governance and control.

“To have autonomy, you need to be able to both understand and control what the agent is doing,” Hathi says.

Click the banner below to keep up with Cisco Live coverage.

 

He views AI observability as an extension of practices enterprises already use to monitor applications, infrastructure and security environments. The difference is that AI systems are inherently less predictable and require tools specifically designed to evaluate behavior at scale.

Organizations also need visibility into the economic impact of AI, Hathi says, including token consumption, operational costs and return on investment. Understanding what AI systems are doing is only part of the equation; businesses also need to know whether those systems are producing meaningful outcomes.

New Cisco and Splunk Capabilities Target AI Visibility

At Cisco Live 2026, Cisco and Splunk are introducing several capabilities designed to help organizations manage AI-driven operations at scale.

Among them is Cisco Data Fabric, designed to help enterprises work with massive amounts of telemetry and machine data across distributed environments. The platform federates data from multiple sources and enables organizations to analyze information without moving it between systems.

LEARN MORE: Cloud and AI drive small business efficiency.

The company is also expanding its agentic security operations center capabilities, which use AI-driven automation to help security teams analyze alerts and respond to threats more efficiently.

Meanwhile, Splunk is extending its AI observability capabilities through its Galileo acquisition. Hathi described Galileo as a platform that helps organizations evaluate agent behavior, monitor performance and enforce guardrails based on observed outcomes.

Taken together, the initiatives reflect what Hathi sees as the next phase of enterprise AI adoption: The challenge is no longer simply deploying AI agents but also ensuring that organizations can see, understand and trust what those agents are doing once they begin acting on their own.

Keep this page bookmarked for articles from the event.

Igor Suka/Getty Images
Close

New Research from CDW on Workplace Friction

Learn how IT leaders are working to build a frictionless enterprise.