The AI productivity trap

Using AI chatbots feels productive at first, but many users later notice problems with memory and skills like writing or coding. Susette Brooks, a marketing director in Philadelphia with an MFA in nonfiction writing, said she felt her writing and thinking decline as she used AI to write emails.

“I felt my writing and thinking decline as it was happening.”

What cognitive offloading means

When you use external tools to reduce the cognitive effort required to do a task, you are engaging in cognitive offloading. Calculators and calendars are simple examples, but AI tools like Claude and ChatGPT can offload much larger tasks like writing term papers or creating work presentations.

Offloading has a cost

Cognitive offloading is not necessarily bad and makes humans productive, but as you outsource more cognitive tasks, you can become forgetful and see skills like writing decline. This process is called de-skilling, and it can affect critical thinking, writing abilities, creative process, and recall.

Gen Z is worried

A Gallup survey with researchers from the University of Pennsylvania found that sixty-eight percent of Gen Z adults in the United States are concerned that offloading too many cognitive tasks is robbing them of important skill-building. About the same percentage worry that AI discourages them from critical thinking and social learning.

The paradox of ease

Even though Gen Z worries about AI's effects, more than half of Gen Z Americans use AI on a weekly basis. Daniel Willingham, a psychology professor at the University of Virginia, says we are hardwired to seek the path of least resistance because cognitive activity is taxing and may not get you to the right solution.

“Our mind is really set up to save you from having to think. Thinking is what you do when you're in trouble.”

The work gets worse too

Derek Salyers, a content creator in Ohio, found that as he used AI to work faster, the quality of his output declined. He said much of what he produced became AI slop, and his brain suffered too because he stopped pausing to actually think about things.

“When you get used to the AI making the decision, you stop pausing to actually think about things.”

Students score higher with AI, lower without it

A study of nearly one thousand high school students found that those using ChatGPT scored forty-eight percent higher on a math test than peers without access. But when tested again without the chatbot, their scores were seventeen percent lower than students who never had ChatGPT access.

The same pattern in text-based work

Researchers at the University of California Irvine found that after ChatGPT's introduction, students spent less time and performed better on text-based questions. But when those same students were later given offline tests without AI help, they performed worse than the control group.

Some skills are more fragile

Evan Risko, a psychology professor at the University of Waterloo who co-authored the foundational paper on cognitive offloading a decade ago, says basic cognitive abilities like working memory and selective attention are quite resilient. But learned behaviors like arithmetic and flying planes require practice to retain.

You might not notice the decline

Brooke Macnamara, a psychology professor at Purdue University, says people may not realize their skills are decaying when using AI tools because the output and their performance with AI is at a high level. They may not know how low their independent performance might be if the AI tool were unavailable.

“They may not realize that if the AI tool were unavailable how low their independent performance might be.”

Be intentional about which tasks

Anjali Singh, a human-AI interaction researcher at the University of Texas at Austin, says the most important step is asking yourself if you should be using AI for a given task at all. Writing is something many people now offload to AI, but doing so means losing the independent thinking and reasoning that goes into expressing your thoughts.

“We are not getting these moments to reflect.”

Use AI for what you already know

A working paper by MIT economist David Autor and colleagues found that giving AI tools to one hundred thirty-three patent lawyers led to different outcomes depending on experience. Younger lawyers saw gains disappear when tools were removed, but experienced lawyers performed better afterward than counterparts who never had AI access.

Augmentation beats automation

A Middlebury College experiment with more than two hundred undergraduates found two types of AI users: augmentation users who used AI to improve their work, and automation users who had AI do the work for them. The augmentation group performed better than everyone else.

“Using AI to deepen or develop one's own thinking on or understanding of a topic, rather than letting AI do all of the thinking, seems like the best way to use these tools.”

Tell AI to teach you

If you are trying to learn something new, you can instruct AI to challenge you and engage with you rather than simply providing answers. ChatGPT has custom instructions in personalization tools, Claude has Instructions for Claude, and companies offer specialized tools like ChatGPT study mode, Claude for Education, and Gemini for Students.

Speed is the enemy of thinking

AI can do work in minutes that takes days manually, which is tempting when workload is heavy. But a Wharton School study found that increased productivity does not necessarily lead to more finished software or happier workers. The faster you move, the more critical thinking you offload to AI.

Make AI interactions collaborative

Slowing down and forcing yourself to do more critical thinking is an easy way to hold onto agency and brain power. One way to do this is to make interactions with AI more collaborative than transactional. For example, ask for five draft options instead of one, then read each and decide which is best.

“Just the simple act of comparing things stimulates a great deal of the critical thinking processes that we kind of care about.”

Introduce friction deliberately

Matt Beane, an associate professor at the University of California Santa Barbara and author of The Skill Code, says letting AI do all the thinking is your worst enemy. You must be very aggressive about how you experiment with it. Evan Risko echoes this: ask for feedback, hints, examples, explanations, advice, and practice questions.

“Ask for feedback, hints, examples, explanations, advice, practice questions.”

Test yourself without AI

The best test for whether you are offloading too much thinking to AI is to stop using it and use your brain instead. Spend a day writing emails without AI to see if you are as fluent as you used to be. Revert to Google for research or build an app with your own code to confirm your skills remain intact.

You can rebuild what you lost

If you find that AI made you worse at something, you are not actually dumber than you were before. You can rebuild your lost skills the old-fashioned way: read, write, do arithmetic. If something requires thinking, go ahead and think.