In Rich’s work advising enterprises on AI transformation at Execution Specialists Group (ESG), a pattern shows up the same way every time. Pilots launch. Licenses get deployed. Training gets scheduled. Leadership gets a dashboard that shows progress. And somewhere between the reporting and the actual work, the question nobody is asking is whether any of it has changed behaviors and business outcomes.

That gap, between what gets reported and what actually changes, is where most AI transformations struggle to succeed. It's not a technology problem. It's not a budget problem. It's a visibility problem layered on top of an execution problem layered on top of a people problem. It compounds faster than most organizations realize. The reporting cycles meant to catch it are slower than the pace of change AI creates.

AI is a means, not a destination. The organizations that find traction aren't the ones with the most ambitious AI strategies. They're the ones that treated AI as an enabling capability for a business strategy that already existed, not as the objective itself. When AI adoption becomes the goal, organizations optimize for having deployed it. Whether it changed anything is a different question, and often one that doesn't get asked until it's expensive to answer.

Transformation happens in behavior, not in deployment. Business value only shows up once adoption actually happens, once human behavior changes and holds long enough to produce a sustained outcome. Everything short of that is activity. The pilots, the licenses, the training completions are inputs to transformation, not evidence of it. The organizations that confuse the two tend to find out the hard way, usually when a second wave of investment is requested for work the first wave was supposed to have accomplished.

This is a workforce shift, not a tooling upgrade. AI implementation is a business process and workforce question before it's a technology question. The emergence of a hybrid human and digital workforce changes how work is planned and executed in ways that don't show up in a deployment timeline or a license count.

Organizational change management (OCM) that arrives late can't see the problem. When OCM is positioned as a downstream function, brought in after the real decisions are made, it inherits the consequences of choices it had no input into. AI accelerates this risk because the pace of change now regularly outstrips the feedback cycles meant to surface it. By the time cultural resistance becomes visible, it's already structural.

How that shapes his approach

AI transformation touches nearly every function of a business: people, process, technology, risk, governance, culture. Claiming fluency across all of it convinces no one of anything in particular. Rich's approach has always been to bring a specific point of view to a specific gap, not a survey of everything that could theoretically apply. The same portfolio discipline that shaped how he built and led PMOs, that prioritization protects an organization's capacity more than raw effort does, applies directly to how he thinks about sequencing AI work now.

The measure he comes back to isn't whether an initiative shipped. It's whether the work actually moved the organization toward the goals it committed to, and whether the organization that delivered it is more capable of doing it again. That's the difference between a career win and a legacy win. His approach to this work is built around the second kind.