Why AI adoption is a people problem, not a technology problem

There is a misconception that industrial companies aren't ready for the AI age, but working with my customers every day, I can confirm it couldn't be further from the truth. The industrial world is the ideal proving ground for AI tools, and the sector has already seen a plethora of innovation built on real-world business needs, creating tools that have transformed operations from engineering to manufacturing.

No, it is not the ambition which is lacking, or indeed the innovation; it's a simple problem with a simple fix: tools need to be designed with people in mind. The system goes live, adoption data comes in, and it turns out that eighty percent of licensed users logged in once and never returned. That pattern repeats across energy, manufacturing, aviation, and field service. This raises an obvious question: if the tool works, and it saves time, why do usage levels reflect this?

The wrong framing produces the wrong outcome

It is natural that organizations assign AI deployment to the IT team; it seems the right fit. However, AI is more than a standard rollout of technology and requires more than a set go-live date and a few training sessions done online with 50 people reading their emails during the session. To understand the lack of adoption, we need to understand what lies behind it, and in the case of AI, it is often trust. By building that confidence alongside the rollout, the tool earns a permanent place in someone's workflow.

In industrial environments, that trust problem is acute. A frontline engineer in a mission-critical setting isn't going to rely on a system they don't understand, especially one that has been positioned as something that monitors their performance rather than supports it. When employees read AI as a surveillance tool, they disengage and work around it; the feedback loop collapses, and the system stops improving.

Organizations that close that gap treat deployment as a change management project from day one. They ask different questions before go-live: what do employees currently understand about how this system works, how has it been positioned internally, and who is responsible for listening to frontline feedback in the first 90 days and actually acting on it?

Positioning determines whether the tool gets used

In industrial settings, the stakes are higher than almost anywhere else. Think about restoring power when temperatures drop to minus 40, or sequencing safety-critical maintenance tasks under serious time pressure. In those moments, a worker needs accurate, contextual information immediately, and is not going to pause to consult a tool they don't trust.

The language used to introduce an AI agent shapes the behavior that follows. Frame it as a productivity monitor, and people will avoid it, or perform surface compliance without genuine engagement. Frame it instead as a collaborative partner, one that takes the institutional knowledge of twenty experienced engineers and puts it in the palm of a first-year technician at precisely the moment they need it, and the dynamic shifts entirely.

That shift has measurable consequences: employees who see the system as a partner give it better feedback, flag errors, and challenge it when something doesn't make sense. That input makes the system more accurate over time and more useful to the next person who needs it. Humanizing the AI agent is what makes the feedback loop function.

What gets lost when organizations rush

No matter if someone is sitting behind a desk or working out in the field, an employee is acutely aware of board-level pressure to deploy AI. The board is right to be concerned with AI deployment and to push for adoption; however, it's important to factor human nature into AI strategy. Never before has technology moved so quickly, and never before has it been more important for leadership to give people training and time to allow uptake to catch up with what the board wants.

Frontline workers are rarely asked what worries them about a new system, or told what happens to their feedback once they give it. Retirements across energy, manufacturing, and aviation are stripping institutional knowledge from workforces at pace, and AI can put what a 25-year veteran knows into the hands of every worker on the ground, with the guardrails to use it safely. That matters more than most organizations acknowledge. Employees who understand that will engage with the system; those handed a login and a slide deck won't.

The channel opportunity hidden in plain sight

This dynamic does not stop at the customer's four walls. Channel partners have an outsized role in how this plays out, and a significant commercial opportunity that most are not yet fully exploiting. Customers rely on their advisors to decide what technology to buy and how to deploy it well. The partner who leads with change management thinking is having a fundamentally different conversation than one leading with feature sets.

In practice, this means asking different questions earlier in the sales cycle: how did your last major software rollout go, and who owns employee experience in this organization? Those questions get at something a standard discovery call never reaches, and they open the door to positioning change management support as a core part of the engagement rather than an afterthought.

Partners who build this capability earn stickier relationships. A customer who goes through a successful AI deployment, one where frontline workers actually use the system, trust it, and give it the feedback that makes it better over time, has experienced something a competitor's feature list cannot easily displace. That is the kind of customer relationship worth winning in the AI era.