In Conversation: Roger Schank on AI’s 2026 Impact on Learning

The Editor: Dr. Schank, it feels like the world is finally catching up to concepts you’ve championed for decades. We’re seeing a boom in AI-driven personalized tutors, sophisticated adaptive learning platforms, and even tools like Notion AI being used to summarize, brainstorm, and structure learning experiences. How do you see these advancements specifically impacting an individual’s learning journey by 2026? What shifts are already underway that you find particularly compelling?

Roger Schank: It’s fascinating, isn’t it? The core idea isn’t new; the computational power and accessibility are. What’s really compelling now is the potential for AI to automate the coaching aspect of learning. Not lecturing, mind you, but coaching. Think about it: traditional education is like being told how to ride a bike without ever getting on one. You read a manual, you listen to a lecture. My argument has always been that you learn by doing, and then reflecting on that doing. Retrieval practice, for instance, isn’t just about recalling information; it’s about actively reconstructing knowledge, finding the gaps, and solidifying pathways in your brain. Spaced repetition works because it forces that retrieval at optimal, increasing intervals, battling the natural decay of memory.

Now, imagine an AI that acts as your perfect sparring partner for these cognitive processes. It doesn’t just present flashcards; it creates scenarios. It observes your struggles, pinpoints conceptual weaknesses, and then designs a unique, personalized “problem” for you to solve. It might say, “You just struggled with conditional logic in Python. Build a simple script that processes user input based on three different conditions, and explain why you chose each branch.” This isn’t just testing; it’s guided application, immediately followed by feedback. The AI automates the Socratic method, but with infinite patience and access to vast knowledge. By 2026, I expect that this kind of personalized, interactive simulation and coaching will be standard for any serious skill acquisition. Forget static courses; we’ll be interacting with dynamic learning environments designed to push us to perform.

The Editor: That’s a powerful vision, essentially turning learning into a highly personalized apprenticeship. But what about the human element? Critics often raise concerns about AI reducing critical thinking or fostering over-reliance. How do we ensure that while leveraging AI for efficiency, we’re also nurturing higher-order skills like creativity, critical analysis, and nuanced problem-solving, rather than inadvertently deskilling learners?

Roger Schank: That’s the crucial point, and it’s why understanding human learning behavior and cognitive load theory is paramount. AI isn’t a substitute for thinking; it’s a tool to amplify thinking. My work on case-based reasoning at Northwestern, for example, highlighted that real learning happens when we’re presented with novel situations and must adapt. If an AI merely feeds you answers, you’re not learning. If it guides you through a complex case study, forcing you to make decisions, analyze outcomes, and then reflect on why those outcomes occurred, then it’s profoundly enhancing your learning.

Consider a professional trying to master strategic negotiation. An AI wouldn’t just give them a rubric. It would simulate a negotiation, complete with a virtual counterpart whose responses adapt to the learner’s tactics. The AI tracks sentiment, identifies logical fallacies in the learner’s arguments, and, post-simulation, provides data-driven feedback: “You missed an opportunity to anchor higher here,” or “Your emotional language escalated the tension unnecessarily at this point.” This builds metacognition – the ability to think about one’s own thinking – because the AI is externalizing and objectifying the learning process. It’s not about spoon-feeding information; it’s about providing infinitely complex problems and immediate, objective performance feedback. The human still has to do the thinking, the creating, the adapting. The AI simply makes the practice richer, more focused, and less prone to the biases of a human instructor who might miss subtle errors or patterns. MIT Media Lab’s work on computational creativity also suggests that AI can serve as a collaborative partner in idea generation, not just a fact-finder, pushing human creativity into new territories. By 2026, we’ll see AI moving from information delivery to sophisticated cognitive scaffolding.

The Editor: So, it’s about AI elevating the practice, not replacing the practitioner. Given this shift, what kind of practical framework should an individual, especially someone looking to future-proof their career, adopt to design an AI-assisted learning plan? How do they move from passive consumption to active, AI-amplified mastery?

Roger Schank: The individual’s role becomes that of a designer of their own learning experiences, not just a recipient. First, identify the core competencies required for your future-proof career. The World Economic Forum’s reports on future skills consistently point to critical thinking, problem-solving, creativity, adaptability, and digital literacy. These are not learned by watching videos.

Second, seek out or build problem-based learning scenarios. If you want to learn advanced data analytics, don’t just take a course; find real-world datasets, define a complex business problem, and then use tools like ChatGPT or specialized AI analytics platforms as your co-pilot. Ask it for data cleaning strategies, interpret error messages, or even brainstorm different model architectures. Use it to do, not just to ask for the answer.

Third, prioritize deliberate practice with AI feedback loops. If you’re learning to write persuasive business proposals, use an AI like Grammarly Business or a custom GPT to critically evaluate your drafts for clarity, tone, and logical flow. Don’t just accept its suggestions; question them. “Why did you suggest this rephrase? What specific principle of persuasion am I violating?” This interactive feedback hones your judgment. I’ve seen countless learners, even in complex fields like software development, using AI not as a crutch but as an always-available debugger and pair programmer. It helps them overcome initial roadblocks and maintain consistency in practice, crucial for skill consolidation.

Fourth, cultivate an experimental mindset. The tools are evolving daily. Try new AI plugins, integrate them into your Notion workspace for project management and learning notes, or experiment with different prompt engineering techniques to get better outputs. This constant experimentation fosters digital adaptability, itself a critical future skill. Recognize that AI is not a static textbook; it’s a dynamic, conversational agent. Your learning journey becomes a continuous process of problem identification, AI-assisted problem-solving, and iterative refinement, always tracking your progress not just by what you know, but by what you can do.

The key is human agency. The AI doesn’t learn for you; it creates an environment where you can learn faster, deeper, and more effectively. The individual who masters leveraging these systems will undoubtedly have a significant competitive edge in the workforce by 2026 and beyond.