IBM Puts Governed Context at the Center of Production AI

In an interview with SiliconANGLE's theCUBE ahead of IBM TechXchange 2026, Clowes said a large language model's reasoning depends on the information available to it, making data delivery an operational concern. "You wouldn't give your accounts receivable work to a nuclear physicist that worked on your energy production side of your business," he said. "So, why would you take a very, very smart model, give it all of the possible information about your business and say, 'Figure out accounts receivable?' You need the right context for the job."

IBM's approach, which Clowes called a "streamhouse architecture," combines streaming, connectivity, processing, governance and querying capabilities to bring together the current state of a business and present it to LLMs. Rather than centralizing all information first, the platform supports preparing usable data products across the environments where business information originates. "The earlier you govern it, the better," Clowes said. "The quicker you produce the high-value data, the better. Then you need to just connect up to all the relevant places."

Delivering current, governed data is only part of the challenge. Agents also need context tailored to their tasks, as different teams may use the same business term to mean different things. Preparing context involves establishing definitions that apply to a specific task, Clowes explained. "You can describe, 'Hey, in this business process, what I mean is X,'" he said. "Then, you can have these governed metrics and these governed datasets that effectively represent the foundation upon which the model will reason." Enterprises can start with high-value tasks and improve individual agents as they gain experience, eventually re-examining how work itself is organized, he added.

As agents take on more responsibility, access to current business information becomes increasingly important. Moving from human-assisted decisions to autonomous operations changes how businesses must deliver information, Clowes said. Agents working continuously need access to operational data without waiting for a person to gather it. "People used to talk about data as the new oil, but what they really meant was [that] data, in some sense, existed and was used by humans to make good decisions," he said. "Businesses were run by humans ... [and] that's how you win." Information may originate on oil rigs, cruise ships or other remote systems, and enterprises must connect those sources into a usable operational view, he said.