The worry of becoming stupid isn't new; it dates back to the time of the Greek philosopher Socrates, who insisted that writing would destroy memory and weaken the mind. Then came calculators, GPS and computers, all of which again raised concerns that we might forget basic mental skills and lose the battle against cognitive offloading. Cognitive offloading is the act of reducing mental effort by using external tools or aids to handle tasks that have historically relied solely on memory or internal processing. A similar sentiment is being observed with the advent of AI.
Subscribe to our newsletter for the latest sci-tech news updates.
A group of researchers set out to investigate an important question: Is AI making us stupid? To find the answer, they reviewed the existing scientific evidence on how using AI to perform mental tasks affects our brains and, particularly, our skills.
The answer is "yes" and "no." What they found was that the impact of AI depends largely on how we use it—whether we use it as a replacement for thinking or as a partner to accomplish our work.
For example, if someone asks an AI to solve a math problem instead of working through it themselves, they miss out on the practice needed to build that skill. Similarly, using AI to summarize information often leads people to put less effort into learning, so they tend to remember less of what they learn as AI is left to handle the hard work of searching, organizing and assessing information.
When using a tutor to fill in knowledge gaps for high school math students, they found that, instead of asking for an answer, their skills developed just as well as students with no AI access at all. The findings are published in Trends in Cognitive Sciences.
How you use it matters
A big concern is that using AI to think for us might make our own minds worse at those tasks when AI isn't around to help. There are two things that need to be considered: skills and basic cognitive abilities.
Skills are the things we learn to do. They combine relevant knowledge with practice, enabling us to perform tasks effectively. Basic mental abilities, however, are different. They are the brain's fundamental, general-purpose functions, such as working memory and attention, that provide the foundation for learning, development and the use of all our specific skills.
When it comes to skills, practice is what builds them. So when we offload that practice to AI, we may never develop the skill in the first place. Even skills we've already mastered begin to decay without regular practice. Practice is one of our strongest defenses against forgetting.
Where the bigger risks lie
There are also concerns about how long-term offloading might affect foundational brain functions over decades, as well as its impact on children during critical stages of development.
In this study, the researchers systematically reviewed existing research from cognitive science, education, medicine and human factors research. They drew on studies of math learning, online information-seeking and AI-assisted medical screening to understand how AI influences the way we think and learn. By focusing on skills, they were able to examine whether we are losing specific how-to knowledge while our basic brainpower remains intact.
Long-term effects remain unclear
The researchers found that these negative effects are possible, but not inevitable. At the same time, it's still too early to know what relying on AI will mean in the long run. Current evidence suggests that everything comes down to how we use AI. Used the right way, AI can boost learning, not by thinking for us, but by working alongside us as a collaborator and a source of feedback.
The real impact of AI on our skills and cognitive abilities won't be clear until long-term studies track how people use it across different tasks and professions.
Written for you by our author Sanjukta Mondal, edited by Sadie Harley, and fact-checked and reviewed by Robert Egan—this article is the result of careful human work. We rely on readers like you to keep independent science journalism alive. If this reporting matters to you, please consider a donation (especially monthly).
More information: Trent N. Cash et al, Is AI making us stupid?, Trends in Cognitive Sciences (2026). DOI: 10.1016/j.tics.2026.06.004
© 2026 Science X Network
This story was originally published on Phys.org.