Every few years, there's a push to simplify. Fewer vendors. Fewer consoles. Fewer late-night calls because some corner of the environment decided to stop working at 2am. The goal is always the same: a leaner, more manageable infrastructure that the team can actually get on top of.
Most AI strategies at the planning stage look roughly the same. There's a line for compute — usually GPU instances, either on-premises or cloud. There's a line for talent, or for the tooling that substitutes for it. There's a governance framework, or at least a placeholder for one. There are pilot projects, proof-of-concept timelines, and a roadmap that gets more speculative the further out it goes.
Public sector IT operates under constraints that the private sector rarely has to think about. Budget cycles that don't flex mid-year. Procurement frameworks that exist for good reasons but add months to every decision. Risk committees that need to be satisfied before anything gets approved. And a culture again, for entirely legitimate reasons that treats caution as a virtue.