AI-first organizations need a fluency baseline.

We provided enterprise AI licenses to everyone, but the way we work at scale hasn't changed.

That is the AI adoption problem in one sentence. Enterprise licenses create access. They do not create an AI-first organization. To change how work gets done, People leaders need an AI fluency and culture baseline that shows readiness, confidence, blockers, team-specific opportunities and progress over time.

The gap is not theoretical. McKinsey found that C-suite leaders estimated only 4% of employees use generative AI for at least 30% of their daily work, when employee self-reports showed the share was three times higher. Leaders were not just underestimating adoption. They were misreading their own organization’s behavior through the wrong lens.

You cannot fix what you cannot see.

If leaders cannot see who is already using AI, where confidence is building, where risk is rising and where work has actually changed, they cannot build an AI strategy. They can only buy tools, send reminders and hope usage reports tell the whole story.

They do not.

Why do usage reports give leaders a false sense of progress?

Usage reports answer a narrow question: Did people access the tool?

They do not answer the questions that matter to a Chief People Officer or head of L&D:

• Do employees know when to use AI and when not to?
• Do managers model useful AI behaviors for their teams?
• Are employees confident enough to apply AI in real work?
• Which teams have the clearest use cases?
• Which teams are stuck because of fear, ambiguity, security concerns or lack of practice?
• Is AI improving the work, or just adding another task?

Without a baseline, leaders are forced to answer strategic questions with anecdotes. The loudest early adopter becomes the proof point. The quietest team becomes invisible. The enterprise dashboard shows licenses assigned, but the People team still cannot see whether AI is becoming part of how decisions get made, customers get served, managers coach or teams communicate.

McKinsey’s broader 2025 workplace AI research makes the same point from another angle: nearly all companies are investing in AI, but only 1% of leaders described their companies as mature, meaning AI was fully integrated into workflows and driving substantial business outcomes. Access moved faster than operating change. That is why the next move cannot be another generic class for everyone. It has to be a clearer read on what your people need next, as shown in McKinsey’s AI workplace report.

What should an AI fluency and culture baseline measure?

A useful AI baseline measures more than technical skill. It turns AI culture into actionable data.

That means looking at five things together:

1. AI fluency. Do employees understand what AI can do, where it helps, where it fails and how to use it responsibly in their roles?
2. Confidence. Are people comfortable experimenting, asking better questions, reviewing AI outputs and applying the results to real work?
3. Cultural readiness. Do teams have the trust, manager support and psychological safety to change habits in public?
4. Blockers. Are employees held back by unclear policy, fear of getting it wrong, tool confusion, data concerns or lack of time?
5. Opportunities. Where can AI change work fastest because the use cases are specific, frequent and tied to business outcomes?

This is where the AI Fluency & Culture Assessment helps. It gives transformation leaders and People teams a shared starting point: an organization-wide baseline, an interactive dashboard and visibility into who needs what.

The output should not be a static report that sits in a folder. It should show where teams are ready to move, where adoption is fragile and where training should be role-specific instead of one-size-fits-all.

AI assessment data can reveal that one team has high curiosity but low confidence. Another team may already use AI every day but lack shared standards. A third may have strong executive pressure but unclear use cases. Those are different problems. They deserve different interventions.

That is the work of strategy, not software.

Why should People leaders own AI transformation?

AI transformation is a workforce transformation. That does not mean People leaders own every tool decision. It means People leaders should own the conditions that make behavior change possible.

The best AI strategy is a People strategy because adoption depends on what people believe, understand, practice and repeat. A Chief People Officer can see the organization as a system: skills, culture, managers, incentives, trust and workload. That system determines whether AI becomes a new way of working or another underused investment.

The workforce stakes are real. The World Economic Forum’s Future of Jobs Report 2025 found that employers expect 39% of workers’ core skills to change by 2030. The same report identifies skills gaps as the top barrier to business transformation, cited by 63% of surveyed employers, with organizational culture and resistance to change cited by 46%. AI adoption sits directly inside that skills and culture challenge.

That is why the People team cannot wait for usage metrics to trickle in from IT. By then, the organization may already have created uneven habits: some employees experimenting in private, some avoiding AI entirely, some moving fast without standards and some managers unsure what good looks like.

A practical AI assessment gives People leaders a cleaner starting point. It helps them tell the business: here is where we are ready, here is where we are stuck, here is where the risk is and here is where targeted learning will move the needle.

How does AI assessment data turn into action?

The point of an AI fluency and culture baseline is not measurement for measurement’s sake. It is prioritization.

Once leaders can see readiness by team, role and blocker, they can stop treating AI adoption like a blanket announcement. They can build a plan that is enjoyable for employees, easy for HR and L&D and effective for the business.

The path usually looks like this:

• Assess. Measure AI fluency, confidence, culture, blockers and opportunities across the organization.
• Develop. Use the results to design live, expert-led classes that match actual needs, not assumptions.
• Practice. Give employees AI simulations, roleplays and real work scenarios so they can build confidence before it counts.
• Embed. Turn learning into prompts, playbooks, manager behaviors, team rituals and clearer expectations.
• Measure. Track adoption, behavior change, time-to-skill and progress over time.

Consider a sales organization preparing managers to coach reps on AI-assisted account planning. The wrong move is a generic AI overview for everyone. The better move is to first assess where managers are confident, where they are skeptical, which workflows are repetitive enough for AI support and where policy confusion is slowing experimentation. From there, live classes can focus on practical account research and better prompts. Simulations can let managers practice coaching a rep through AI-generated insights. Reporting can show whether managers are applying the skill afterward.

That is a different experience for employees. It respects their time. It connects to their work. It gives People teams proof beyond attendance.

If you are building from scratch, our guide to an AI fluency assessment strategy outlines how to move from assessment to a clear people plan in under two weeks. If you are already rolling out training, the next step is making sure your learning builds company-wide AI fluency that changes behavior, not just awareness, as we cover in building AI fluency at scale.

What happens when you stop guessing?

When you start with AI assessment data, the conversation changes.

Instead of “How many people logged in?” you can ask, “Which teams are ready to change how work gets done?”

Instead of “Who attended training?” you can ask, “Who applied the skill this week?”

Instead of “Do we need more AI classes?” you can ask, “Which intervention will remove the next blocker?”

That shift matters because AI adoption is not a communications campaign. It is a behavior-change program. People need clarity, practice, manager support and evidence that the new way of working is worth the effort.

Electives is built for that work. Our AI adoption training blends live classes, AI simulations and analytics so organizations can assess, develop and change how work gets done. Across our AI training, 98% of learners apply their new AI skills within one week. That number matters because application is the goal. In practice, it can look like a manager using AI to prepare for a coaching conversation, a sales team turning account research into a stronger plan or an operations team replacing a manual first draft with a better AI-assisted workflow.

That is the difference between access and transformation.

If your organization has already invested in AI licenses but still cannot see how work is changing, do not start with another generic rollout. Start with an AI Fluency & Culture Assessment. Build the baseline. Find the blockers. Then give every team the most relevant path to becoming AI-first.