The Dual Reality at Davos

I spent part of last week reading the articles that came out of Davos. The World Economic Forum had convened its usual panels on technology and scale, and one phrase from the coverage stuck with me: what they called the "messy middle between demonstration and broad adoption."¹

It's a useful idea. Technologies don't stall because they fail. They stall because the institutions trying to adopt them can't absorb the change fast enough. The panels were thoughtful. Saudi Aramco reported roughly six billion dollars in realized value from technology integration, about half from AI. The diagnosis was careful, credible, and aimed at the right problem.

And then I thought about the person who was never there.

The diagnosis is correct. The patient isn't in the building.

Two Realities, One Organization

The same week those panels were happening, a survey of 5,000 white-collar workers told you where that messy middle actually lives. Forty percent of employees said AI saves them no time at all. Only two percent of executives said the same.²

That's not a rounding error. That's two populations describing different realities in the same organization.

This got me thinking about what I keep seeing with clients. The leaders who champion AI have lived through the uncertainty, debated trade-offs, earned their clarity. By announcement day, the strategy feels overdue. For the rest of the organization, it arrives as interruption. Another tool, another login, another thing to manage on top of everything that was already full.

A separate survey found 31 percent of employees actively sabotaging their company's AI efforts.³ Not because they're Luddites. Because the sanctioned tools don't solve problems employees actually have.

Meanwhile, 60 percent use unsanctioned shadow AI tools just to meet deadlines.⁴ They're not resisting the technology. They're routing around the strategy.

The dashboard says adoption. The sticky note says otherwise.

The Translation Layer

You can perfectly diagnose the absorption problem from a stage in Switzerland and still have no mechanism to reach the person on the office floor who just logged into the AI tool, let it run in the background, and kept working the old way.

The Davos panel isn't wrong about anything. That's what makes the dual reality so revealing. The institutional language, how ideas spread, who funds the patience, how to retrain the workforce, none of it is incorrect. But it lives at a different altitude than the employee doing quiet arithmetic: does this tool make my day shorter, or does it add a new thing I have to manage?

I wrote about GM's seat bracket a few weeks ago, a part their AI designed that the factory couldn't build. That was a gap between what AI could imagine and what organizations could produce. This is a different pattern. The gap isn't between AI and the organization anymore. It's between two groups of people inside the same organization, both telling the truth about their own experience, neither able to see the other's.

What sits between them is a translation layer. Not more technology, not another strategy deck. Someone who can convert institutional ambition into workflow reality. Someone who asks what stops when the new thing starts. When 31 percent of your workforce is quietly working around your AI strategy, you don't have a technology problem. You have a workflow problem. The tool works. It just doesn't fit the work.

No amount of panels will fix that. But someone willing to walk between both floors might.

---

References

Footnotes

1. World Economic Forum. (2026). "How we can deploy innovation and technology at scale and responsibly." WEF ↩

2. Section survey of 5,000 white-collar workers, reported by Mashable. (2026). "Does AI save time? Executives say yes, employees say no." Mashable ↩

3. Writer & Workplace Intelligence. (2026). "GenAI's Silent Rebellion." WebProNews ↩

4. BlackFog. (2026). "Corporate workers lean on shadow AI to enhance speed." Cybersecurity Dive ↩