.svg)
AI agents are full of promise, but still remain short on value.
Today, there is no shortage of different agentic tools for organizations to adopt. However, enterprise leaders are frustrated at the lack of real-world returns from many of these solutions. AI investments might be generating productivity and efficiency gains in pockets. But ultimately, the technology isn't delivering impact where it matters: the bottom-line.
At the same time, companies are fearful of new security risks and the loss of important intellectual property. As a result, businesses are hesitant to give AI agents too much access. They want senior-level performance from an “employee” that may only be able to work in a single application. This distrust of the technology threatens to hinder progress at a critical juncture.
The answer? AI and orchestration, together: “It’s a really powerful combination to be able to move super fast and deploy these agents across the enterprise,” Samsara Chief Information Officer Stephen Franchetti told TheCube at World of Workato 2025.
Most organizations haven’t even yet witnessed a true AI agent in action. Too many are flashier chatbots masquerading as ground-breaking new technology. They’re incorrect, and skip critical steps. They also can inject new vulnerablities, and potentially even hinder operational excellence.
Typically, these so-called “AI agents” are only operating in segments of the workflows that drive operations. Order-to-cash, supplier management, employee and customer lifecycles, IT help desk; these are complex processes involving multiple underlying technologies. AI agents have to be able to work across all the places where business runs: applications, databases, unstructured documents, and more.
The relevant context from these end sources then helps AI agents level-up the workflows that define the business and make it unique. Dropping three days off of an order-to-invoice process can significantly impact cash flow, for example. Or when IT delivery happens in seconds, not days, companies can move faster to tackle opportunities.
Accurate and armed with the right process context, AI agents can start to anticipate problems and opportunities. But trust remains key to the shift from defensive to proactive.
