AI Literacy Is Not AI Capability: What Education Must Build Next

The First Layer Is Awareness

Every capability journey begins somewhere.

People need to understand:

• what AI is;
• what different AI systems can do;
• where AI is being used;
• what its limitations are;
• what risks may exist;
• and how it is changing the environment around them.

This is the foundation of AI literacy.

Without basic understanding, people can struggle to distinguish meaningful AI use from inaccurate claims, inappropriate applications or unreliable outputs.

AI literacy therefore matters. But literacy is a starting point. It should not become the endpoint.

Knowing AI Is Different From Using AI

The next stage is practical use.

A learner may know how to:

• write a prompt;
• generate an image;
• summarize a document;
• ask an AI system a question;
• generate an outline;
• or use an AI-enabled application.

These are useful capabilities.

But practical operation alone does not establish whether the learner understands:

• why the tool should be used;
• whether the output is reliable;
• what information was used;
• what may be missing;
• whether another method would be better;
• or what responsibility remains with the person using it.

The difference is similar to the distinction between operating a tool and understanding how to use the tool appropriately.

Education needs to address both.

From AI Literacy to AI Capability

A useful progression is:

Each stage represents a different level of capability.

The progression matters because capability is not demonstrated by reaching the tool-use stage alone.

The Assessment Question

This creates a difficult question for education.

A learner completing an AI tutorial demonstrates exposure. A learner successfully generating an output demonstrates application. Neither necessarily demonstrates judgment.

Meaningful capability requires evidence that a learner can:

• understand a problem;
• determine whether AI is appropriate;
• select an appropriate approach;
• use AI effectively;
• examine the resulting output;
• identify limitations or errors;
• improve the result;
• explain relevant decisions;
• and take responsibility for the final application.

This changes the role of assessment. Instead of asking only “Can the learner use the AI tool?”, education increasingly needs to ask:

That is a capability question.

AI-Assisted Learning Is Not the Same as AI-Substituted Learning

The distinction becomes particularly important inside education itself.

AI can support learning by helping learners:

• explore unfamiliar concepts;
• generate examples;
• receive explanations;
• compare perspectives;
• practice communication;
• test ideas;
• receive feedback;
• and work through problems.

But there is a difference between using AI to strengthen learning and using AI to substitute for learning.

If a system produces the answer before the learner has developed the ability to understand the problem, the resulting output may demonstrate AI access rather than learner capability.

The objective should therefore not be to remove AI from learning. Nor should it be to allow AI to perform every intellectually demanding part of the learning process.

The more important question is how AI can be integrated while preserving understanding, inquiry, judgment, reasoning, creativity, responsibility and independent capability.

The Educator’s Capability Changes Too

The transition does not apply only to students. Educators also need capability to work effectively in an AI-integrated environment.

That includes understanding:

• where AI can support teaching;
• where AI may introduce risk;
• how AI-generated material should be evaluated;
• how assessment practices may need to evolve;
• how learners should be taught responsible AI use;
• and where human teaching and judgment remain essential.

This means AI readiness in education cannot be reduced to giving educators access to AI tools. It requires the capability to design, evaluate and manage AI-supported learning environments.

The educator therefore moves beyond being simply a user of AI tools. The role increasingly includes being a guide for how learners understand and work with AI.

Education Must Move From Completion to Capability

Traditional education systems often organize progression around:

An AI-integrated environment creates a stronger need to examine:

This does not make qualifications irrelevant. It changes what qualifications and learning systems need to demonstrate.

A certificate can establish that a person completed a defined pathway. Capability requires stronger evidence of what that person can understand, apply, evaluate and execute.

This is particularly important as AI makes the production of certain forms of output easier. When generating an answer, image, presentation or piece of code becomes increasingly accessible, the educational question shifts.

The scarce capability may no longer be simply producing the output. It may be knowing what to produce, why it matters, whether it is correct, how it should be improved and what responsibility accompanies its use.

Information Literacy Becomes More Important, Not Less

AI capability also intersects with information literacy.

AI systems can produce information quickly, but speed does not guarantee accuracy. Learners therefore need to develop the ability to:

• question information;
• examine sources;
• recognize uncertainty;
• identify unsupported claims;
• compare evidence;
• distinguish fact from interpretation;
• and understand the limitations of generated outputs.

This makes critical evaluation a central component of AI capability. The more accessible generated information becomes, the more important the ability to evaluate information becomes.

Responsible AI Is Part of Capability

AI capability also includes responsibility. People need to understand questions involving:

• privacy;
• data;
• intellectual property;
• bias;
• misinformation;
• security;
• appropriate use;
• human oversight;
• and accountability.

Responsible AI should therefore not exist as a separate theoretical topic disconnected from practical use. It should become part of the capability itself.

A person who can operate an AI system but cannot recognize when its use is inappropriate has developed only part of the required capability.

The Post-Tool Capability Layer

As AI tools become easier to access, the value of simply knowing how to operate a particular tool may decline. Tools change. Interfaces change. Models change. Workflows change. New systems appear.

The underlying capability therefore needs to extend beyond any particular platform. This is the post-tool capability layer. It includes:

• problem framing;
• critical thinking;
• systems thinking;
• judgment;
• adaptability;
• information literacy;
• responsible technology use;
• communication;
• continuous learning;
• and execution.

A learner who understands these capabilities can continue developing even as specific AI tools change. That makes capability more durable than tool familiarity.

What Education Must Build Next

The next phase of AI education therefore requires a broader architecture. It should help learners move from:

This is not a rejection of AI literacy. It is the logical progression beyond it.

AI literacy provides the foundation. AI capability builds the ability to operate.

From Literacy to Readiness

The distinction can ultimately be expressed simply.

Education has an opportunity to build all three. The challenge is ensuring that the first layer does not become the final destination.

As AI becomes increasingly embedded across education, professional work and society, the question will not only be whether people have access to AI. It will be whether they have developed the capability to use that access with understanding, judgment, adaptability and responsibility.

Being Topper’s Perspective

Being Topper approaches this transition through the broader lens of Digital & AI Capability .

The institution’s capability-first approach treats AI readiness as more than familiarity with tools. It connects understanding, application, evaluation, adaptability, responsible technology use and execution.

The broader progression is:

Within education, this means connecting learning with application and moving beyond tool proficiency toward capability that remains useful as technologies and working environments continue to change.