A framework for ensuring student AI proficiency

From engineering and finance to marketing and product design, employees are now expected to use AI to analyse information, generate ideas and support decision-making. So, graduates entering the workforce will be expected to collaborate with these tools as part of routine workflows.

My own perspective on AI literacy comes from teaching computer science students at two stages of engagement with these tools. I work with students in foundational courses who are encountering these technologies for the first time. At the same time, I collaborate with students on state-of-the-art research projects in AI. Moving between these contexts makes one thing clear: preparing students for an AI-enabled workplace requires more than technical exposure. They need practical ways to understand the technology, experiment with it and develop judgement about how it should be used.

Over the past few semesters, I have structured my teaching around a framework that helps students build that capability: demystify, use and reflect.

Demystify the technology

Many students arrive with strong opinions about AI but only a partial understanding of how these systems work. Some see them as nearly magical tools that can produce answers instantly. Others dismiss them as unreliable or assume they are only useful for technical specialists.

Demystifying AI begins with explaining the basic ideas behind large language models (LLMs) and related systems. We show students how these models are trained, what kinds of data they rely on and why their outputs can sometimes appear confident even when they are incorrect.

• Spotlight guide: Boosting AI literacy across your institution
• ‘Employability is more than getting a job’
• What if the undergraduate journey were a four-year internship?

At their core, LLMs are mathematical models trained on enormous data sets that predict the next token in a sequence of words based on probabilities learned from data. One of my students exclaimed after learning the basics: “Oh, it’s just math!”, illustrating the moment when understanding the underlying statistics transformed AI from something mysterious into something approachable.

That moment also opened up a useful conversation: if AI is built on data patterns then the quality of that data, the design of the system and the context in which it is used all matter. Understanding this helps students see AI as a tool that is fallible and thus must be evaluated and improved.

Use the tools in real workflows

Once students have a clearer understanding of the technology, the next step is to experiment with it directly. I encourage students to explore how AI tools can support tasks such as brainstorming ideas, summarising information, analysing data sets or assisting with code development. The goal is not simply to produce answers faster but to help students learn how AI can become part of a broader workflow.

For example, students might compare their own analysis of a dataset with AI-generated output. They can then assess what the tool does well, where it falls short and how human expertise improves the final result. This process mirrors what many graduates will encounter in professional settings. AI tools are rarely used in isolation; they are incorporated into collaborative processes where people must evaluate, refine and build on what the systems produce.

Reflect on when and how AI should be used

The final step is reflection. Students need opportunities to think critically about when AI tools are helpful and when they may introduce new risks or limitations. Reflection activities ask students to evaluate the accuracy of outputs, consider potential biases in data, assess how AI use might influence decision-making and evaluate their own AI-assisted processes. These conversations help students develop sound judgement about where automation adds value and where human insight remains essential.

In practice, this stage often leads to deeper discussions about responsibility, transparency and trust in AI-assisted work. I highly recommend encouraging these discussions.

Building interdisciplinary AI literacy

This approach to teaching AI literacy is also shaping a new interdisciplinary AI minor at Virginia Tech. The programme will bring together students from fields including computer science, engineering, art, business and the social sciences. The goal is to recognise that AI will influence nearly every discipline, and students should be able to work effectively with these systems.

Our intent is that students in the minor will learn not only how AI systems function but how they can be applied in professional contexts. They will examine questions about implementation, collaboration and ethical decision-making alongside the technical foundations.

Preparing students for an AI-enabled workplace

As AI continues to evolve, universities will need to help students develop both technical fluency and critical awareness. The goal is to prepare them to engage with these tools thoughtfully and productively.

When students understand how AI works, experiment with it in realistic contexts and develop the judgement to evaluate its outputs, they are better prepared for a workplace where human creativity and machine capability will increasingly work side by side.

Margaret Ellis is professor of practice in the department of computer science at Virginia Tech. 

If you would like advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the Campus newsletter.