Excellent piece! Can you suggest any of the recent research you mention that shows the negative correlation between AI usage and critical thinking? I find that checks out intuitively and would be interested to read more. I find that one of the most interesting and impactful places for considering the impact of "unchecked cognitive offloading" is in the workplace. There seems to be a growing mainstream line that goes that people should upskill to have those abilities that AI can't replace. On the surface that makes sense, but it does call into question the metacognitive skills and preparedness of how people who are already classified today as "educated". Also, I worked with assessments for many years so what you say on that point is resonant for me. Arguably, the top skill for success on assessments has long been a kind of suspension of disbelief that allows one to gear up to perform on the assessment. A company call GPT Zero did some work on checking AI usage across entire courses or departments in a university setting with the idea being not to spot AI usage as cheating but to encourage evaluation of those assessments that seem to point students most toward AI tools.
What a rich comment—thank you! The GPT Zero example is incredible. What a time-saver for teachers. I imagine GPT Zero could filter out the tests that drive inappropriate cognitive offloading, decompose the assumptions of the test, lift those assumptions to a more sophisticated level, and design better tasks for both instruction and assessment.
Regarding the research on AI and critical thinking, let me pull together some studies for you. Several recent works have examined this relationship:
The most comprehensive is Gerlich's 2025 study (with 666 participants across diverse backgrounds), which found "a significant negative correlation between frequent AI tool usage and critical thinking abilities, mediated by increased cognitive offloading." Younger participants showed higher AI dependence and lower critical thinking scores.
The Microsoft study I referenced (January 2025) surveyed 319 knowledge workers, revealing that "higher confidence in GenAI is associated with less critical thinking, while higher self-confidence is associated with more critical thinking."
Dr. Ahmed's engineering research found that among 285 students, "68.9% of laziness and 27.7% of the loss of decision-making skills were the result of artificial intelligence use."
Interestingly, a 2023 study on language learners suggests that students with previous AI experience actually demonstrated higher critical thinking in AI-supported environments—pointing to the importance of metacognitive training.
As Gerlich notes, "AI isn't inherently bad for our cognitive abilities—like any tool, it needs to be used correctly."
Key Studies:
- Gerlich, M. (2025). "AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking" - https://www.mdpi.com/2075-4698/15/1/6
Excellent piece! Can you suggest any of the recent research you mention that shows the negative correlation between AI usage and critical thinking? I find that checks out intuitively and would be interested to read more. I find that one of the most interesting and impactful places for considering the impact of "unchecked cognitive offloading" is in the workplace. There seems to be a growing mainstream line that goes that people should upskill to have those abilities that AI can't replace. On the surface that makes sense, but it does call into question the metacognitive skills and preparedness of how people who are already classified today as "educated". Also, I worked with assessments for many years so what you say on that point is resonant for me. Arguably, the top skill for success on assessments has long been a kind of suspension of disbelief that allows one to gear up to perform on the assessment. A company call GPT Zero did some work on checking AI usage across entire courses or departments in a university setting with the idea being not to spot AI usage as cheating but to encourage evaluation of those assessments that seem to point students most toward AI tools.
What a rich comment—thank you! The GPT Zero example is incredible. What a time-saver for teachers. I imagine GPT Zero could filter out the tests that drive inappropriate cognitive offloading, decompose the assumptions of the test, lift those assumptions to a more sophisticated level, and design better tasks for both instruction and assessment.
Regarding the research on AI and critical thinking, let me pull together some studies for you. Several recent works have examined this relationship:
The most comprehensive is Gerlich's 2025 study (with 666 participants across diverse backgrounds), which found "a significant negative correlation between frequent AI tool usage and critical thinking abilities, mediated by increased cognitive offloading." Younger participants showed higher AI dependence and lower critical thinking scores.
The Microsoft study I referenced (January 2025) surveyed 319 knowledge workers, revealing that "higher confidence in GenAI is associated with less critical thinking, while higher self-confidence is associated with more critical thinking."
Dr. Ahmed's engineering research found that among 285 students, "68.9% of laziness and 27.7% of the loss of decision-making skills were the result of artificial intelligence use."
Interestingly, a 2023 study on language learners suggests that students with previous AI experience actually demonstrated higher critical thinking in AI-supported environments—pointing to the importance of metacognitive training.
As Gerlich notes, "AI isn't inherently bad for our cognitive abilities—like any tool, it needs to be used correctly."
Key Studies:
- Gerlich, M. (2025). "AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking" - https://www.mdpi.com/2075-4698/15/1/6
- Microsoft Research (2025). "The Impact of Generative AI on Critical Thinking" - https://www.microsoft.com/en-us/research/publication/the-impact-of-generative-ai-on-critical-thinking-self-reported-reductions-in-cognitive-effort-and-confidence-effects-from-a-survey-of-knowledge-workers/
Thank you for the references!