Companies that cut their workforce to fund AI adoption are not getting the returns they expected. That is not speculation. Research is now making it increasingly clear that automation driven job cuts are failing to generate the financial results business leaders banked on. The productivity gains look convincing in a presentation. The reality on the ground is a lot messier.
The Numbers Are Not Matching the Narrative
The logic seemed straightforward enough. Remove headcount, reduce labour costs, deploy AI to cover the gap, watch margins improve. For a lot of companies, that equation has not held up.
What the emerging research is pointing to is a pattern that anyone who has been through a major technology transition will recognise. When you remove people to fund a technology investment, you do not just reduce a cost line. You remove the institutional knowledge, the informal processes, the contextual understanding, and the relationship networks that made the work function in the first place. AI tools cannot inherit any of that. They can only work with what they are given.
The result, in many cases, is that the AI ends up operating in a knowledge vacuum. It processes information efficiently but without the depth of understanding that the people who left had built over years. The output is faster. It is not always better. And when something goes wrong, the people who would have known how to fix it are no longer there.
What Gets Lost When the People Leave
There is a concept in organisational management called tacit knowledge. It refers to the things people know how to do that they cannot fully articulate in a document or a process manual. A senior customer service rep who knows exactly how to handle a frustrated long term client. A logistics coordinator who has learned through experience which supplier to call when the usual one falls through. A finance analyst who understands the context behind a number that looks wrong on paper.
This kind of knowledge does not sit in a system. It sits in people. It is built through years of doing the work, making mistakes, and adjusting. When businesses cut the people who hold that knowledge to fund AI tools, they are not just reducing headcount. They are deleting institutional memory that took years to accumulate and cannot be easily rebuilt.
AI tools in their current form are genuinely impressive at processing structured information, identifying patterns at scale, drafting content, and automating repeatable tasks. They are not good at replacing the judgment, context, and human understanding that experienced team members carry. The businesses that are struggling with AI returns right now are often the ones that did not understand where that line is.
The Businesses Outperforming Are Doing Something Different
The research thread that the social post above links to points to a finding that sits at the centre of everything XSIV is built around. The businesses achieving genuine AI returns are not the ones replacing people with technology. They are the ones upgrading their people alongside their technology.
That distinction matters more than most business leaders currently appreciate.
When AI is introduced into a team that has been prepared for it, the people already understand the context, the exceptions, the nuances, and the edge cases. The AI handles the volume and the repeatable tasks. The people handle the judgment calls, the relationships, and the situations that require genuine understanding. The combination produces outcomes that neither the AI nor the people could achieve alone.
When AI is introduced to replace the people who held that understanding, the technology is left to operate without the context it needs to perform at its best. The efficiency gains are real in the short term. The strategic performance gaps tend to show up later, and they are harder to diagnose because they are not visible in a straightforward cost metric.
This is not a people first argument dressed up as a business case. It is what the data is starting to show. The businesses treating AI as a workforce replacement strategy are underperforming relative to the businesses treating it as a workforce amplification strategy.
What This Means for Business Leaders Watching from the Outside
If you are a business owner or leader in Australia watching large companies make this mistake at scale, the relevant question is not whether their model will work. The evidence is accumulating that it largely does not. The relevant question is what your own AI roadmap says about your team.
Most AI roadmaps are technology documents. They outline which tools will be adopted, which workflows will be automated, what the implementation timeline looks like, and what the projected efficiency gains are. Very few of them have a section on how the people in those workflows will be supported through the transition, what they will be trained on, how their roles will evolve, and what they need to feel confident rather than threatened.
That gap is where most AI adoption failures live. Not in the technology. In the people and process side that the technology roadmap did not account for.
The businesses that are getting this right are the ones where the leadership team asked a different set of questions before they started. Not just which tools to buy, but which team members need support to adapt, which workflows carry tacit knowledge that needs to be documented before automation, and how to bring the team on the journey rather than presenting AI as something happening to them.
The Window Is Still Open
The businesses starting their AI readiness work now are not behind. They are right on time. The companies that rushed in early without a people strategy are the ones now dealing with the consequences that the research is documenting. There is something to learn from watching that play out at scale before you move.
What the data is pointing toward is not a reason to slow down on AI adoption. It is a reason to be deliberate about how you approach it. The technology is accessible. The pricing is coming down. The tools are improving. None of that is a sustainable competitive advantage on its own. The advantage is a team that understands why things are changing, has been given time and support to adapt, and knows how to work alongside AI rather than feel threatened by it.
That is not built by cutting people and replacing them with software. It is built by investing in people and giving them better tools to work with. The businesses that understand that distinction are the ones that will look back in five years and recognise they made the right call.
If your AI roadmap is mostly a technology document right now, it is worth asking what the people section looks like. If there is not one, that is probably the most important thing to address before anything else.
Get in touch at www.xsiv.au/#form