Educators aren’t the only ones thinking about the relationship between generative AI and critical thinking. A recent study by Lee et al. (2025) from Carnegie Mellon University and others from Microsoft Research surveyed 319 knowledge workers about their experiences using GenAI in their jobs. The researchers found that while confidence in GenAI was associated with less critical thinking, self-confidence was associated with more critical thinking. This study included a diverse sample of professionals and 936 real-world examples of AI use. There are some intriguing parallels for educators considering how to integrate AI into the curriculum while developing critical thinking abilities in our students.

Key Findings

Lee et al. (2025) asked participants to describe specific instances of using GenAI at work and then report on how the technology related to tasks like organizing their thoughts and evaluating new information. Participants also described the effects of GenAI on their critical thinking. To reveal trends, these responses were analyzed alongside demographic data and opinions about trust in GenAI.

Lee and colleagues found a multi-step process that workers employed when “thinking with AI.” This involved: 1) setting a goal and creating prompts for the AI tool, 2) evaluating if the AI’s output was adequate, and 3) choosing and modifying the output to fit the needs. This process underscores the importance of prompt engineering and careful evaluation, skills that matter in both the workplace and the classroom.

The study found that confidence was strongly related to how knowledge workers approached GenAI. Respondents who had high confidence that GenAI was competent were less likely to prioritize critical thinking, which could lead them to depend too much on the technology. Respondents who were confident in themselves were more likely to rely on their critical thinking.

Implications for Educators and Students

This has implications for educators. Scholarship suggests that students and graduates approach novel tasks with a lack of confidence that causes them to seek exemplars (Baird et al. 2022). Anecdotal evidence suggests that students are often turning to AI tools to generate text, solve problems, and conduct research. It’s important to consider how students’ confidence levels might influence willingness to try critical thinking. Are students with high confidence in AI blindly accepting its outputs, or are they actively evaluating and refining them?

Previous research supports the notion that students already recognize the importance of balancing AI-generated insights with their own critical thinking skills (Córdova et al. 2024). Studies have shown that students know about the risks of over-reliance on AI to the detriment of developing their own problem-solving abilities and that they understand the need to expand on and verify AI’s output.

Furthermore, educators are increasingly recognizing the need to design learning experiences that incorporate AI in a developmental way (Deschenes and McMahon 2014). This includes teaching students essential AI skills such as prompt engineering, ethical considerations, and advanced analysis and critical thinking. By integrating AI into the curriculum, we can equip students with the skills they need to navigate an AI-driven world while making plain the relationship to their own critical thinking abilities.

Conclusion

The Lee et al. (2025) study reveals that the use of GenAI tools in the workplace shifts the focus from information gathering to information verification, from problem-solving to AI response integration, and from doing tasks to supervising tasks. The authors suggest that GenAI tools themselves should be designed to support knowledge workers’ critical thinking by addressing their awareness, motivation, and ability barriers. This could help students, too, and educators could be a part of the process of tool design.

References

Baird, Neil, Alena Kasparkova, Stephen Macharia, and Amanda Sturgill. 2022. ““What One Learns in College Only Makes Sense When Practicing It at Work”: How Early-Career Alumni Evaluate Writing Success.” In Writing Beyond the University: Preparing Lifelong Learners for Lifewide Writing, edited by Julia Bleakney, Jessie L. Moore, and Paula Rosinski. Elon University Center for Engaged Learning. https://doi.org/10.36284/celelon.oa5.10.

Córdova, Pamela, Alberto Grájeda, Juan Pablo Córdova, Alejandro Vargas-Sánchez, Johnny Burgos, and Alberto Sanjinés. 2024. “Leveraging AI Tools in Finance Education: Exploring Student Perceptions, Emotional Reactions and Educator Experiences.” Cogent Education 11 (1). https://doi.org/10.1080/2331186X.2024.2431885.

Deschenes, Amy, and Meg McMahon. 2024. “A Survey on Student Use of Generative AI Chatbots for Academic Research.” Evidence Based Library and Information Practice 19 (2):2-22. https://doi.org/10.18438/eblip30512.

Hao-Ping (Hank) Lee, Advait Sarkar, Lev Tankelevitch, Ian Drosos, Sean Rintel, Richard Banks, and Nicholas Wilson. 2025. “The Impact of Generative AI on Critical Thinking: Self-Reported Reductions in Cognitive Effort and Confidence Effects From a Survey of Knowledge Workers.” CHI Conference on Human Factors in Computing Systems (CHI ’25). https://www.microsoft.com/en-us/research/uploads/prod/2025/01/lee_2025_ai_critical_thinking_survey.pdf.

About the Author­­­­   

Amanda Sturgill, Associate Professor of Journalism at Elon University, is a 2024-2026 CEL Scholar focusing on the intersection of artificial intelligence (AI) and engaged learning in higher education. Connect with her at asturgil@elon.edu.   

How to Cite this Post   

Sturgill, Amanda. 2025. “Unlocking the Link Between Generative AI Confidence and Critical Thinking Skills.” Center for Engaged Learning (Blog). Elon University, March 25, 2025. https://www.centerforengagedlearning.org/unlocking-the-link-between-generative-ai-confidence-and-critical-thinking-skills.