So you’ve identified an AI skills gap – maybe from an organizational assessment, or from an AI tool that scans workflows and flags areas for development. Either way, you know that your team needs support to get the greatest ROI from AI.
Before you jump straight into building training content and best practices guides, stop. Ask a harder question first: is what you identified actually a skills gap? Or is it something else entirely?
At Data Society, we see a pattern that gets misdiagnosed constantly. Organizations identify a gap, fund a training program, and then watch adoption stay flat. Not because the training was bad, but because the real problem was never a capability gap in the first place.
We call it the translation gap.
THE TRANSLATION GAP: DESIGN VS. EXPERIENCE
The translation gap is the distance between what the employer designs (the strategy, the tools, the workflows, the guidance) and what the employee actually experiences day to day.
When that gap exists, employees are not failing to learn. They are failing to see how any of this connects to the work in front of them. And no amount of training content closes that distance.
“I almost see it as a grid, where as the Tech maturity increases, you need your People maturity to increase as well to keep coming up those steps.”
— Catie Maillard, Global Head of People, Data Society Group
The translation gap shows up when an organization has invested in a robust AI infrastructure, but the rollout experience for employees feels disconnected from their actual workflows. When there is no bridge between what leadership envisioned and what people encounter at their desks, training alone cannot build that bridge.
The first intervention, before you design a single course or build a single best practices guide, is change management. That means communication, context-setting, leadership alignment, and honest dialogue about what is actually changing and why. It means asking whether your people understand where AI fits into their role — not just how to use the tool. It means making sure they have access to your learning tools quickly and easily and have time to dig into training if they need them.
If the answer is no, you do not have a skills gap. You have a translation gap. And the fix is different.
THE GAP IS ALWAYS MOVING
Here is what makes this harder: even once you close the gap, the target moves.
AI tools are not static. They evolve and use cases evolve – often faster than organizations can absorb them. The capability your team built six months ago may already be trailing behind what the tools now make possible. On the flip side, the people who were early adopters may now be ahead of where your infrastructure can actually take them.
“You’ve built this incredibly complicated piping system, but nobody knows how to use it, and there’s no water flowing through it.”
— Catie Maillard, Global Head of People, Data Society Group
This is not a failure of implementation. It is the nature of the moment. The organizations that navigate it well are the ones that treat AI readiness as a continuous investment — not a one-time initiative.
That means both sides of the equation have to keep moving forward together. Technology infrastructure and people readiness are not two separate workstreams that converge at a launch date. They are two tracks on the same rail, and if one advances without the other, you stall.
WHAT THE AI SKILLS GAP ACTUALLY LOOKS LIKE
The AI skills gap is not one gap. It is two, and they compound each other — and the translation gap sits between them.
On one side: organizations invest heavily in AI tools, platforms, and infrastructure, but their people do not have the skills, context, or confidence to use them. The technology sits underutilized. ROI does not materialize.
On the other side: employees who are enthusiastic about redesigning their workflows, but they do not have access to the right tools, or the tools are not integrated into anything meaningful. Their effort goes nowhere.
“You’ve got this great infrastructure set up, but no one’s using it. The other side is you have people who are working on redesigning their processes, but they don’t have access to a tool. In which case, they’re not seeing the gains that they need there either.”
— Catie Maillard, Global Head of People, Data Society Group
Both gaps need to close together. And as the tools evolve, that closing is never finished. For a clear breakdown of how these two tracks work in practice, see AI Upskilling vs. AI Implementation: Why Enterprises Need Both.
WHY THE PEOPLE SIDE GETS UNDERINVESTED
The technology side of AI investment is easy to justify. There are vendors, demos, and clear deliverables. People investment is harder to sell internally — especially when the ROI is diffuse and slow to appear.
But underinvesting in people does not just slow things down. It actively undermines the technology investment you already made.
“It will be like putting a Jaguar engine in a golf cart. It’s not going to have the intended impact.”
— Catie Maillard, Global Head of People, Data Society Group
And because the tools keep advancing, delaying People investment does not just mean you are behind today. It means the gap compounds. Every quarter you wait, the distance between your infrastructure and your workforce’s ability to use it grows wider.
THE TWO CATEGORIES OF SKILLS YOUR ORGANIZATION ACTUALLY NEEDS
Technical skills include understanding how AI tools work, how to prompt effectively, how to evaluate AI output critically, and how to make decisions about when AI adds value. These are learnable and increasingly essential across every function — not just technical roles.
Human skills include critical thinking, judgment, oversight, communication, and the ability to redesign processes rather than just layer AI on top of existing ones.
“We need to be increasing our technical knowledge, increasing our data governance, data readiness, the way everything is connected in our systems, at the same time as we’re upskilling and advancing and redesigning work for our people. Because without one, you are going to fall short.”
— Catie Maillard, Global Head of People, Data Society Group
Neither category is a one-and-done investment. Both have to grow in parallel with the tools themselves. That is not an argument for constantly reinventing your training infrastructure — it is an argument for building the conditions where continuous learning is the norm, not the exception.
HOW THE AI SKILLS GAP CONNECTS TO RETENTION AND CULTURE
The AI skills gap has a retention dimension that often gets overlooked. Employees who feel underprepared for the tools their organization expects them to use do not just underperform — they disengage. And as the pace of AI change accelerates, that feeling of being left behind compounds quickly.
For a closer look at the workforce retention dimension, see CIO Dive: Retention Deficit: Upskilling Gives Tech Workers a Path Forward
The productivity case is equally strong. Forbes: Why Data Literacy Increases Productivity Among Employees offers external validation for the investment in people development.
HOW TO DIAGNOSE WHAT YOU’RE ACTUALLY DEALING WITH
Before designing any intervention, the right starting question is: do we have a skills gap, a translation gap, or both?
A real AI skills assessment looks at:
• Current tool adoption and how employees are actually using AI day to day
• Workflow design — does AI fit into how work actually happens, or does it sit outside it?
• Employee confidence levels
• Cultural signals around experimentation and psychological safety
• The gap between what leadership thinks is happening and what employees are experiencing
That last point is where the translation gap usually lives. If your people know the tools exist but cannot connect them to their work, change management comes before curriculum.
The 2025 AI Readiness Report from Data Society provides a benchmark for where organizations typically are on the readiness spectrum.
THE BOTTOM LINE
“We need both of these to go hand in hand.”
— Catie Maillard, Global Head of People, Data Society Group
Technology and people readiness have to advance together — not because it sounds good, but because the alternative is that neither investment delivers. And because the tools are not going to stop evolving, neither can your approach to closing the gap.
That means every time you are evaluating a new AI tool or workflow, you are also asking: do our people have what they need to make this work? And more specifically: do they understand what this means for them?
If the answer is unclear, that is not a training gap. That is a translation gap. Start there.
READY TO CLOSE YOUR AI SKILLS GAP?
If your organization’s AI skills gap feels like a strategy problem — or a translation problem — not a training problem, you are probably right. Data Society helps teams close the gap between technology investment and people readiness through customized AI upskilling, change management advisory, and organizational readiness work.
Connect with our team to learn what that looks like for your organization:
https://datasociety.com/contact/
RELATED RESOURCES FROM DATA SOCIETY
AI Upskilling vs. AI Implementation: Why Enterprises Need Both:
https://datasociety.com/ai-upskilling-vs-ai-implementation/
The 2025 AI Readiness Report:
https://datasociety.com/the-2025-ai-readiness-report-insights-to-build-your-2026-strategy/
CIO Dive: Retention Deficit: Upskilling Gives Tech Workers a Path Forward:
https://datasociety.com/cio-dive-retention-deficit-upskilling-gives-tech-workers-a-path-forward/
Forbes: Why Data Literacy Increases Productivity Among Employees:
https://datasociety.com/forbes-why-data-literacy-increases-productivity-among-employees/
Frequently Asked Questions
The AI skills gap refers to the mismatch between the AI capabilities organizations need and the skills their employees currently have. It encompasses both technical skills and softer competencies like critical thinking, AI oversight, and workflow redesign. It also includes what we call the translation gap — the distance between how leadership has designed the AI strategy and what employees actually experience.
The translation gap is the distance between design and experience — between what an organization intends for its AI implementation and what employees actually encounter in their day-to-day work. It is not a capability gap. It is a communication and change management gap, and it requires different interventions.
The AI skills gap is caused by the rapid pace of AI tool development, underinvestment in people development relative to technology investment, a tendency to treat AI training as a compliance exercise, and cultural conditions that discourage experimentation and open learning. It is also caused by treating the gap as static — when in reality, it moves every time the tools advance.
Closing the AI skills gap starts with diagnosing what kind of gap you actually have. If it is a translation gap, the first intervention is change management — not curriculum design. From there, closing the gap requires a parallel approach: investing in technical AI training while also building the human capabilities that make AI use meaningful. Because the tools keep evolving, this is not a one-time initiative. It requires building conditions for continuous learning.
It is a strategy problem. The tools exist. The issue is whether organizations have built the human infrastructure to use them — and whether they have done the translation work to connect technology investment to day-to-day employee experience.
What happens if an organization ignores the AI skills gap?
Organizations that ignore the AI skills gap tend to see AI initiatives stall, produce inconsistent quality, and fail to deliver the productivity gains that justified the investment. And because the tools keep advancing, the gap compounds. Over time, it also becomes a talent issue — employees who feel underprepared are more likely to disengage or leave.
