Designing for the Thinking We Can't Outsource

Throughout this series, we've explored a variety of instructional design strategies for teaching in an AI-rich world. We've talked about redesigning assignments, making student thinking visible, building reflection into learning, teaching verification and evaluation skills, creating richer discussions, and writing AI expectations that connect technology use to learning goals.

At first glance, these may seem like separate instructional techniques. I don't think they are. I think they're all trying to accomplish the same thing. They're helping us identify, and protect, the thinking students still need to do for themselves.

That realization has gradually changed how I think about AI in education. When generative AI first became widely available, much of the conversation centered on what students could produce. Could they write the paper? Complete the discussion post? Generate the presentation? Solve the problem?

Those questions were understandable. But over time, I found myself asking a different question. What thinking was the assignment actually designed to develop? That question changes everything. Because once we ask it, AI becomes less of a threat to the assignment and more of a lens for examining the learning itself.

Throughout this series, we've repeatedly returned to that idea. When we redesign assignments, we aren't simply changing activities. We're creating opportunities for students to make decisions, justify their reasoning, and apply knowledge in meaningful ways.

When we make thinking visible, we're looking beyond the finished product to better understand the reasoning, decisions, and connections that produced it. When we build reflection into learning, we're helping students examine their own thinking, recognize what they understand, and identify where they still have room to grow. When we teach verification and evaluation skills, we're asking students to question information instead of simply accepting it. To weigh evidence. To recognize uncertainty. To exercise judgment. When we design better discussions, we're creating opportunities for students to encounter perspectives that challenge their assumptions and refine their understanding through dialogue. And when we explain AI expectations in terms of learning goals, we're helping students understand not just what the rules are, but why certain kinds of thinking belong to them.

Different strategies. One purpose. Protecting the thinking that matters. That doesn't mean students shouldn't use AI.

Quite the opposite.

Professionals already work alongside powerful tools. Engineers use simulation software. Designers use digital modeling tools. Physicians use diagnostic systems. Programmers use AI-assisted coding environments. Learning has never been about refusing to use tools.

It's always been about developing the judgment required to use those tools well.

I believe the same is true for AI. Students should absolutely learn how to use AI effectively. They should learn how to ask better questions. How to evaluate AI-generated information. How to recognize its limitations. How to integrate it into their workflow responsibly.

Those are important skills. But they are not the whole story. The deeper question is this: What thinking should remain the student's responsibility?

Because while AI can generate explanations, summarize information, organize ideas, suggest solutions, and even produce convincing arguments, there are still forms of thinking that education should intentionally cultivate.

The ability to evaluate competing evidence. The ability to make decisions when there is no single correct answer. The ability to recognize assumptions. The ability to connect ideas across contexts. The ability to reflect on one's own understanding. The ability to exercise sound judgment under uncertainty.

Those capacities have always mattered. AI simply makes them easier to see.

In many ways, I think that's been the hidden lesson throughout this entire series. AI hasn't diminished the importance of learning. It has clarified what learning has always been about. Not simply acquiring information. Developing judgment. Not simply producing answers. Making sense of ideas. Not simply completing assignments. Becoming capable of thinking independently in a world filled with increasingly capable tools.

That is why I don't believe the future of higher education lies in competing with AI or trying to prevent students from using it. I think our future lies in becoming much more intentional about identifying the thinking that belongs to students and designing learning experiences that help them practice that thinking.

Because the most valuable outcomes of education have never been the artifacts students submit. They've always been the thinking those experiences were designed to develop. And perhaps that is the most important lesson AI is teaching us.

Not what can be automated. But what shouldn't be.

Series 1: AI Is Exposing Existing Problems ✓ Completed
Series 2: What We Do About It ✓ Completed

Over the next series, I'd like to explore that final idea more deeply. What kinds of thinking should never be outsourced? And how should higher education change when AI becomes a permanent partner in learning rather than a temporary disruption?

I believe that may be one of the most important questions education will face over the coming decade.

Current Post (8 of 8): Small Teaching Changes That Increase Engagement

Next Up: SERIES 3 – Cultivating Human Thinking

First Post: Judgement