By DataCenterKnowledge
Publication Date: 2026-05-05 12:22:00
At IBM Think 2026, IBM introduced what it calls an ‘AI operating model’ built around data, agents, automation, and hybrid infrastructure – its blueprint for moving enterprises from isolated AI projects to coordinated systems embedded in core business processes.
The push lands as enterprises confront growing infrastructure constraints tied to AI deployment, including data gravity, latency, and rising power demands inside data centers. Rather than centralizing workloads in hyperscale environments, IBM is positioning hybrid and on-premises architectures as necessary to run AI closer to enterprise data.
Arvind Krishna, IBM chairman and CEO, framed the shift during a Monday media call in blunt terms: “Over 70% of all the data is still sitting inside the enterprise. You have to act where the data is.”
From Projects to Systems
IBM’s model frames a transition from narrow use cases to enterprise-wide systems. “The companies that are doing well will get 40% enterprise…
