Embrace It, Don't Shame It: Using AI to Enhance Student Learning and Problem Solving

When ChatGPT launched in late 2022, universities around the world had essentially the same reaction. Panic.

Within weeks, institutions were issuing emergency guidance. Some banned AI tools outright. Others rushed to adopt detection software, and suddenly every assignment submission was suspect. AI could clearly be misused, and universities had to decide how to respond. Most institutions chose restriction, and some brought back handwritten exams.

We went the other direction.

From Caution to Conviction

We spent 2023 using these tools in our own work (writing, coding, data analysis), and they changed how we worked. As thinking tools, they helped us iterate faster and get past mechanical bottlenecks. We wrote about this shift in 95% of My Work Happens in VS Code.

But we also saw the other side. Students submitting AI-generated essays with no understanding of the content. Detection tools flagging non-native English speakers as “AI-written” while missing actual AI output.

By 2024, we’d started actively showing students how to prompt effectively and verify the output without giving up their own voice and judgement. Students who learned to use the tools well became better thinkers.

Now, in 2026, we’ve built an entire course around this philosophy. The course is called Practical AI for Behavioural Science, and it requires AI use. Every student uses ChatGPT, Claude, GitHub Copilot, and Gemini throughout the semester. We assess their understanding and their judgement in using these tools, and the AI helps them build both faster.

The Stigma That Remains

In research, the conversation has moved from “should we use these tools?” to “how should we use them responsibly?” But in teaching, the stigma persists. Many institutions still treat AI use in student work as something to be prevented, detected, and punished. Under a ban, students get no practice in prompting or in checking AI output, though they will need both after they graduate. Hours spent on detection come out of teaching time.

AI as a Thinking Partner

The biggest misconception driving the fear of AI in education is that it gives students the answers. It can, if you let it. Whether it does depends on how students are taught to use it. When a student types “write me an essay about cognitive dissonance” into ChatGPT, they learn nothing. When a student types “I’m arguing that cognitive dissonance theory underestimates the role of social context. What are the three strongest counterarguments to my position, and which papers support them?”, they’re doing real intellectual work, with the AI as a sparring partner.

Students have to be taught how. With a framework and practice, they learn to use AI the way a good researcher uses a knowledgeable colleague, to pressure-test ideas and iterate toward something better than either could produce alone, as we describe in our prompt engineering guide.

Critical Thinking Gets Stronger

The original fear was that AI would erode critical thinking. In our teaching since 2024, we’ve seen the opposite. When students verify AI output, checking whether the citations actually exist, confirming the statistics make sense, evaluating whether the reasoning holds up, they develop verification habits they never had before. Because students can’t trust AI blindly, its imperfections push them to think critically in a way a textbook rarely does.

We structure this with the LLM Problem-Solving Loop, two nested loops in which the student makes the decisions and checks the AI’s work.

There’s an outer research loop (plan, execute, evaluate, document) that you’d follow regardless of whether AI existed. Wrapped inside it is an inner AI loop. You give the AI context and make it propose a plan before it generates anything, then you verify everything it returns against what you know. The inner loop is expected to run two to five times per task. Refining a prompt after a bad result is one of the skills the course assesses.

Never use LLM output without verification. You are responsible for every claim you take from it.

Learning to Ask Better Questions

Working with AI forces students to spell out what they want. A vague prompt gets a vague answer. To get something useful, you have to define your problem precisely and specify what “good” looks like before you start. These are exactly the skills we try to teach in research methods courses and thesis supervision, except now there’s an immediate, tangible feedback loop. Write a bad prompt, get a bad result, figure out why, and improve it. We explore this idea more in Prompt Engineering Is the Skill Nobody Teaches.

Removing Bottlenecks Without Losing the Thinking

Our course is designed so that psychology students with no coding background build machine learning pipelines within the first few weeks. AI coding assistants break down the technical barriers that would otherwise make this impossible. The AI handles the syntax, and students focus on modelling and interpretation questions: Is this the right model for this question? What does this result mean? What are the limitations?

Beyond coding, AI can clear mechanical bottlenecks in any discipline, so students spend more time deciding whether an approach fits the question and what a result means.

Transparency Over Surveillance

If we want students to be honest about their AI use, we have to go first. The course materials themselves were designed and developed with Claude, ChatGPT, GitHub Copilot, and Gemini, and this is stated openly. We even code all our lecture slides in HTML with AI assistants. Requiring students to disclose their AI use while pretending we don’t use the same tools is hypocritical.

For the written assignment, students submit their complete, unedited chat histories alongside their work, because the process is part of the assessment. Forty percent of the rubric grades the quality of the AI interaction: how well they prompted, whether they iterated, whether they pushed back when the AI was wrong, whether they verified claims. A student who engages critically and builds their own argument from verified claims shows the judgement in using AI that the course aims to develop.

A Policy Isn’t a Pedagogy

Many institutions are responding to AI by writing policies: “You may use AI tools, but the work must be your own.” This sounds reasonable. In practice, it’s almost useless. Students have no framework for what “the work must be your own” means when an AI helped produce it. The ambiguity makes them anxious, and many end up using AI without telling anyone.

Educators should teach AI use as a skill, starting with a structured framework like the LLM Problem-Solving Loop and examples of good and bad AI interaction. They should also grade the process along with the product, as our rubric does by giving forty percent of the marks to the AI interaction. You don’t need to teach a whole course on AI to do this. The framework can be introduced in a single lecture.

This Is Just the Beginning

This is the first time we’re running this course in its current form. Semester 1, 2026, started this past week. So far we can only describe the course design. We’ll report results in two follow-up posts, one mid-semester and one at the end, with reflections on what worked and what we’d change. The course repository is open-source on GitHub, with materials released week by week. If you’re an educator thinking about how to handle AI in your teaching, follow along and take what’s useful.

Let’s see how it goes.

---

Michael Richardson Professor, School of Psychological Sciences Faculty of Medicine, Health and Human Sciences Macquarie University

Rachel W. Kallen Professor, School of Psychological Sciences Faculty of Medicine, Health and Human Sciences Macquarie University

Ayeh Alhasan Dr, School of Psychological Sciences Faculty of Medicine, Health and Human Sciences Macquarie University

---

AI Disclosure: We wrote this article with help from AI tools, including Claude. The ideas and the course design described here are our own. The AI helped us draft and edit the text. We use AI throughout our research and teaching, and we encourage others to do the same. Using AI well is a skill to be learned openly.

The AI models used here, like all current LLMs, were trained on vast quantities of text written by others, largely without explicit consent. Every output they produce draws on the ideas and language of countless researchers and writers. Their labour makes these tools possible, and we want to acknowledge it even where no individual can be credited.