AI Washing Is Costing Executives More Than They Realize

The Mistake Nobody Names

Are your AI investments showing up in dashboards but not in your business results? That gap has a name now.

On July 24, 2026, Matt Domo published a piece in Entrepreneur that put a precise label on one of the most expensive blind spots in enterprise AI strategy. Domo — who spent years helping build AWS and now guides companies through AI transformations as CEO of FifthVantage — calls it AI washing. Not the consumer-facing version of that phrase, where brands slap "AI-powered" on products that don't warrant it. Something more specific, and more damaging: the internal version, where leadership celebrates activity metrics while actual business performance quietly erodes.

The pattern looks like this. Adoption rates climb. Response times drop. The AI pilot gets a standing ovation at the all-hands. Meanwhile, customer satisfaction is drifting, retention is softening, and nobody is connecting those two things to the AI rollout because the rollout looks successful by every metric the team agreed to track.

Domo's argument is that those metrics were the wrong ones to begin with. Measuring how fast your system responds tells you nothing about whether customers are better served. Measuring how many employees use the tool tells you nothing about whether they are doing more meaningful work. The mistake is not deploying AI. The mistake is confusing deployment with transformation.

Why Activity Feels Like Progress

Part of what makes AI washing so hard to diagnose is that the metrics fueling it feel genuinely useful. Pilot completion rates, tool adoption curves, response time improvements — these are real numbers, produced by real systems, reported in real dashboards. They are not invented. They are just answering the wrong question.

The psychological pull here is easy to underestimate. Outcome metrics — customer retention, employee satisfaction, decision quality — are slow, noisy, and contested. Did retention drop because of the AI rollout, or because of a pricing change, or because a competitor made a move in Q2? That ambiguity is uncomfortable to sit with in a quarterly review. Adoption rates have no such problem. Either people used the tool or they did not. The number is clean, it is defensible, and it fits neatly into a slide.

Matt Domo's framing makes this explicit: celebrating faster response times while customer satisfaction declines is not a measurement error. It is a prioritization error. The team chose metrics that were easy to report over metrics that were honest to examine.

That tradeoff compounds quickly. The longer leadership tracks the wrong indicators, the more resources flow toward initiatives that look productive but are not moving the outcomes that actually matter. The dashboard stays green. The business problem stays unsolved.

The Three Questions That Cut Through

Domo's diagnostic framework does not ask whether your AI initiative is running. It asks whether it is working — and it applies three questions to determine the difference.

The first: does this improve the customer experience? Not the internal process behind it. Not the speed at which your team completes the task. The actual experience of the person on the other end. Domo points out that faster response times and declining customer satisfaction can coexist — and often do. If the AI initiative cannot show movement on a customer outcome, the question of whether it belongs in the portfolio at all becomes legitimate.

The second: does this help employees do more meaningful work? Not just different work, or faster work, but work that requires more of their judgment, creativity, and expertise. If employees are spending their reclaimed hours on lower-value tasks — or if the tool has simply made their day more fragmented — that is a signal worth acting on.

The third: does this enable better decisions for leaders? More data is not the same as better decisions. Domo's framing here is specific — the question is whether the AI output is actually changing what leaders decide to do, and whether those decisions are producing better results.

Apply all three as filters, not a checklist. An initiative that passes one and fails the other two is still a problem worth naming.

Accountability Over Ambiguity

Knowing the right questions is only useful if someone is responsible for answering them.

Domo's practical corrective starts with an audit — not of tool deployments or pilot counts, but of whether your AI investments have produced any measurable change in actual outcomes. Customer retention. Employee satisfaction. Decision quality. If you cannot draw a line between the investment and movement on one of those three, that is the audit result. The initiative has not transformed anything yet.

The second corrective is structural. Assign a single accountable owner to each AI initiative, tied to an outcome — not a process. Shared ownership across a steering committee is not ownership. When customer satisfaction dips and the AI rollout is a plausible contributing factor, accountability diffused across five stakeholders produces one outcome: a meeting where everyone explains why it was not their area.

The third is the one most organizations skip because it feels slow. Talk to customers and employees before you measure anything else. Not surveys deployed after the fact. Direct conversations, early, while there is still time to change course. Customers will tell you whether the experience changed. Employees will tell you whether the work got better or just different. Both conversations are faster and cheaper than six months of dashboard data pointing in the wrong direction.