Stanford Graduate School of Business research has found that tech workers lead AI adoption at 60% frequent usage. But here is the part that matters more than the headline number. Growth has stalled. After surging between 2024 and 2025, adoption rates plateaued across organisations. The tools did not get worse. The people and organisational factors simply did not keep pace with the rollout.
The First Wave Always Looks Like Success
Every organisation that rolls out an AI tool goes through the same initial surge. The curious pick it up immediately. The early movers experiment, find wins, and share them. Usage numbers climb. Leadership points to the adoption curve and calls it a success.
Then it stops.
Not because the tool stopped working. Not because the team lost interest. It stops because the first wave of adoption was never really adoption. It was experimentation by the people who were already predisposed to try new things. The rest of the team, the majority of the team, was watching. And what they saw was a tool being rolled out, not a change being led.
This is the pattern that plays out in business after business. The Stanford GSB research is not an anomaly. It is a data point that confirms what many business owners are quietly experiencing right now. The tool is in the business, but the business has not actually changed.
Rolling Out a Tool Is Not the Same as Building Capability
There is a meaningful difference between giving your team access to an AI tool and building a team that knows how to use it well, trusts it, and has had time to genuinely adapt their workflows around it.
Most AI rollouts focus on the first part. The procurement decision gets made. The licences go out. Someone runs a lunch and learn. A few power users emerge. And then the expectation is that adoption will spread naturally from there.
It rarely does. Because the people who did not pick it up in the first wave are not slow or resistant for no reason. They are busy. They are uncertain about whether using the tool correctly is actually in their job description. They are worried about making a mistake with something they do not fully understand. Some of them are carrying a quiet fear that the more useful this tool becomes, the less useful they become.
None of those things get solved by access. They get solved by leadership, communication, and time.
The Human and Organisational Gap Is Where Adoption Dies
Industry data consistently points to human and organisational factors as the primary reason AI adoption efforts fail, not technical ones. The Stanford GSB research reinforces this. The stall is not happening because the technology reached its ceiling. It is happening because organisations hit a ceiling on the human side and did not have a plan for what comes next.
That ceiling looks different in every business, but the common threads are recognisable. There is no clear expectation from leadership about how AI should be used and by whom. There is no feedback loop for people to share what is working and what is not. There is no psychological safety to experiment and get it wrong. And there is almost never a structured conversation about what AI means for people’s roles and what the business is doing to make sure that transition is fair and supported.
When those things are missing, the early movers carry on using the tool and the rest of the team quietly reverts to what they already knew. The adoption curve flatlines. Leadership scratches their head wondering why the investment is not delivering the productivity gains they were promised.
The answer is almost always the same. The technology was sorted. The people side was not.
What the Businesses Getting This Right Are Actually Doing
The organisations that move past the first wave plateau are not doing anything exotic. They are doing the unglamorous work that most businesses skip because it takes longer and does not show up in a dashboard.
They are talking to their teams before the tool arrives, not after. They are being honest about why the change is happening and what it means for people’s roles. They are identifying who in the team needs more support and giving them that support without making them feel behind. They are building shared expectations about how AI fits into the work, not leaving it up to individuals to figure out on their own.
They are also being patient. Not passive, patient. There is a difference. Real adoption takes longer than a quarter. People need time to build confidence with a new way of working, and that confidence does not come from a single training session. It comes from doing, making mistakes in a safe environment, and gradually finding the moments where the tool genuinely makes their work better.
The businesses that get this right are the ones that treat AI adoption as a people project that happens to involve technology, not a technology project that involves some change management at the end.
The Real Question for Your Business Right Now
If your organisation rolled out an AI tool in the last 12 to 18 months and you are seeing the usage curve flatten, the question worth sitting with is not which tool to try next. It is what your team’s actual experience of the transition has been.
Have they been told clearly what is expected of them? Have they been given the time and support to build real capability rather than just surface familiarity? Do they feel like the change is being done with them or to them? Do they understand what their role looks like as AI becomes more embedded in the work?
Those are not soft questions. They are the operational questions that determine whether your AI investment compounds or stalls. The Stanford GSB research shows the stall is common. It does not have to be permanent. But moving past it requires being honest about where the gap actually is, and in most cases, the gap is on the people side.
The technology was always the easy part.
Get in touch at www.xsiv.au/#form