This is part of the Connecting Ed Tech Leaders with Cutting-Edge Research Series.

AI doesn’t have to be the problem—it can be part of the solution if we design it right and teach students to use it intentionally. Rather than viewing tools like ChatGPT as replacements for student thinking, educators can use them to strengthen critical reasoning and creativity. Critical thinking is more than logic or correctness—it’s the process of deciding what to believe by evaluating evidence, testing assumptions, and refining ideas. When used thoughtfully, AI can help students engage in these deeper processes rather than bypass them.

chatgpt image nov 5, 2025, 11 56 16 am Research by Stanford’s Hari Subramonyam highlights how AI can scaffold, rather than shortcut, the thinking process. His team developed Script & Shift, a writing platform designed to guide learners through idea development, organization, and audience awareness. Instead of producing text for students, the tool uses interactive “Writer’s Friends” such as Idea Ivy and Structure Sam to prompt reflection, revision, and creativity. In a study of 90 students, those using Script & Shift showed stronger critical thinking and knowledge transformation—the ability to synthesize and reframe ideas—than peers using ChatGPT or no AI support. They also reported a greater sense of agency over their writing and thinking.

For K-12 technology leaders, this research underscores the importance of selecting and promoting AI tools that support how students think, not just what they produce. Banning AI is not the solution; instead, districts should teach students to use it for brainstorming, clarifying ideas, and seeking feedback. Purpose-built, scaffolded tools can nurture curiosity, reflection, and self-directed learning—helping students move beyond copying information to constructing understanding. The future of education lies in using AI not to automate thought, but to amplify it.

For more on this topic, listen to the Stanford Graduate School of Education’s School’s In podcast episode “Beyond ChatGPT: Critical Thinking in the Age of AI”

hari Author and Researcher:  Hari Subramonyam,  Assistant Professor at Stanford University (CA)

 

Published on December 9, 2025

Hari Subramonyam is an Assistant Professor at the Graduate School of Education and Computer Science (by courtesy) and a Faculty Fellow at the Institute for Human-Centered AI at Stanford University. His research applies a design and cognition lens to AI in education, focusing on how AI tools and methods can advance the learning sciences to protect and augment learner cognition and agency. Drawing from cognitive psychology, human–computer interaction, and learning theory, his work explores how AI systems can co-create knowledge with learners, support conceptual understanding, and foster more equitable, human-centered forms of learning.

CoSN is vendor neutral and does not endorse products or services. Any mention of a specific solution is for contextual purposes.