Organizations that have invested in data literacy often assume they are halfway to AI literacy. The logic sounds right: both are about working with information, both need some comfort with numbers, and both are taught through training programs. So AI gets added as a module to the existing curriculum.
Our position is that these are related but different skills, and treating them as one hides the gap that matters most. We score them separately, for each role. A team can be strong in one and weak in the other, and the difference between the two, role by role, is usually the most useful finding a literacy program produces.
Two skills, often treated as one
Data literacy is reading, questioning and acting on numbers. Can someone read a chart and say what it does not show? Do they check where a surprising figure came from before repeating it? Do they notice when a metric has quietly changed definition between quarters?
AI literacy is knowing what an AI system can and cannot do, using it safely, and checking what it returns. Can someone explain why a model can be fluent and wrong at the same time? Would they verify an AI-generated figure or citation before it reaches a client? Do they know what happens to information they paste into a tool?
Both involve judgment about evidence. But they ask people to be skeptical of different things, in different ways.
Why they diverge
A dashboard and an AI assistant fail differently. A dashboard that is wrong usually looks wrong in a way a trained reader can spot: a truncated axis, a figure that disagrees with another report, a number that has not refreshed. Data literacy trains people to look for those signals.
An AI system that is wrong often looks exactly like one that is right. The tone is the same, the formatting is the same, and a fabricated citation reads as smoothly as a real one. A confident tone carries no information about accuracy. The habits that catch a bad chart do not catch a plausible paragraph.
The reverse also happens. People who use AI tools fluently every day can be weak at questioning a metric, because the tools answer so readily that the question of where the number came from never comes up.
What goes wrong when you blend them
- One score hides the gap. A team that is strong on data and weak on AI averages out to "moderate", and the program teaches both at a moderate level. The real risk, uncritical use of AI output, goes unaddressed.
- The wrong people get the wrong course. Analysts sit through data basics they already know; executives get a tool demo when they needed to learn how to judge an AI proposal.
- Responsible use becomes an afterthought. Safe use of AI, what never goes into a prompt and when a person must decide, ends up as a short compliance module at the end instead of part of the skill.
- Completion is counted, use is ignored. Everyone finished the course. Nobody changed how they work.
What each role needs, through both lenses
Literacy is not one level for everyone. A director and an analyst need different things, and both tune out of a course built for the other. Looking at each role through both lenses makes the target concrete:
| Role | Data literacy | AI literacy |
|---|---|---|
| Executives | Read a dashboard critically and ask for the definition behind a number | Judge an AI proposal: value, readiness, risk and who is accountable |
| Managers | Use team data to decide, and know when it is not good enough | Decide where AI belongs in a workflow and where a person must stay |
| Analysts | Build on governed data and document what they did | Use AI tools on real work, and verify what comes back |
| Frontline | Enter and read data correctly at the point of work | Use approved tools safely, and know what never goes into a prompt |
Executives usually go first. People adopt what their leaders visibly understand and ask about, and the executive view decides what the rest of the program should prioritize.
Knowing what not to do is part of literacy
AI literacy includes responsible use: who may use which tools for what, what stays out of a prompt, when a person must make the decision, and when to escalate. That belongs inside every track, taught on real cases, not bolted on as a final module.
Regulators have started to address AI literacy directly. The EU AI Act contains an AI literacy provision in Article 4, which, as amended by the Digital Omnibus on AI, asks providers and deployers of AI systems to take measures to support the AI literacy of their staff. Our guides to the EU AI Act and Illinois AI law set out what each covers.
A completion certificate is not compliance, and neither is a learning dashboard. What an organization can show is what was done: baselines, role competencies, records and results. Whether that meets a particular obligation is a legal question.
How to measure each
Measure before and after, separately for each skill, and judge the program by what people do differently rather than by attendance.
| Stage | Data literacy | AI literacy |
|---|---|---|
| Baseline | Reading, questioning and acting on numbers, scored by group | Limits, safe use and verification, scored by group |
| Practice | Exercises on the dashboards and data people already use | Exercises on approved tools, inside your policies |
| After | Fewer reconciliation arguments, questions answered from governed data | Appropriate use, outputs verified, fewer preventable mistakes |
| Evidence | Baselines, competencies, records and results, kept and handed over | |
We go further into measurement in measuring literacy by use, not attendance.
Where to start
- Baseline each group on both skills, scored separately, before anything is taught.
- Look for the gap between the two scores in each role. That is where the program should focus.
- Start with the executive team, so the program's priorities come from the people who fund it.
- Build tracks by role, practiced on your own tools, data and policies.
- Put responsible use inside every track.
- Measure use and judgment afterwards, and keep the evidence.
This article is not legal advice. Whether a literacy program meets a particular obligation depends on the facts and the law that applies to you.
