Sedentary behavior is a useful comparison for understanding what could happen to our thinking skills in the age of AI. No single missed workout ruins your fitness. One week of missed workouts barely matters. But a year of consistently skipping exercise changes your baseline – your cardiovascular capacity drops, your muscle mass shrinks, your metabolic health gets worse. The change is invisible day to day and obvious year over year.
The same thing may happen to your thinking in the age of generative AI. This post is an analogy about practice and skill, not a claim about clinical cognitive decline; if you notice changes in your memory or thinking, see a clinician. No study yet shows that routine AI use causes a measurable loss of judgment. The closest evidence is correlational, such as a survey-and-interview study of 666 people, which reported a negative correlation between frequent AI tool use and critical-thinking scores, mediated by cognitive offloading [1], which can show association but not cause.
The Missed Rep
Every time you hand off a judgment call to AI that you could have made yourself, you skip one rep of mental exercise. One rep doesn’t matter. The question you would have worked through, the choice you would have made, the tradeoff you would have weighed – these are like a single squat for your brain.
One missed squat doesn’t change your body. One skipped judgment call doesn’t change your mind. So don’t worry about any single one of these.
But hundreds of missed reps over a year could change your baseline.
Here’s an illustrative calculation, using hypothetical numbers rather than measured ones. If you use AI for ten judgment calls a workday – “draft this email,” “summarize this document,” “suggest options for this problem” – that’s roughly 2,500 outsourced judgment calls a year. Even if half of those genuinely make sense to hand off, the other 1,250 are missed reps. In the analogy, that’s like skipping every workout for a year.
The Neural Mechanism
The mechanism proposed here is “use-dependent plasticity”: brain circuits that get used often tend to grow stronger, and circuits that get used rarely tend to grow weaker [2]. That is a well-established general principle about how the brain adapts to experience, but it does not by itself show that delegating decisions to AI weakens adult reasoning.
The skills that may be most at risk from AI outsourcing are the ones involved in the hardest parts of thinking: evaluation (weighing options against several criteria), synthesis (pulling together information from different sources), and taste (developing and applying your own quality standards). These are complex skills, and they plausibly need regular use to stay sharp.
When you skip the evaluation step and just accept the AI’s first answer, you aren’t saving time. In the analogy, you’re choosing not to exercise your evaluation skills. One choice like that doesn’t matter. The concern is the buildup.
The comparison to physical exercise holds up for another reason: it works both ways. Just as someone who’s out of shape can rebuild cardiovascular fitness through consistent training, someone who’s grown too dependent on AI can likely rebuild rusty mental skills through deliberate practice. The difference is that a decline in skill is hard to see – you may not feel yourself getting shallower the way you feel yourself getting winded climbing stairs.
The Hidden Tax
This cost is hidden because your environment adjusts to your declining capacity. If your evaluation skills weaken, you may not notice worse reasoning. You may notice that you trust AI output more, or that you second-guess yourself less. It can feel like confidence – but it may mean your standards have dropped.
This may be the most dangerous part: it can feel like progress. You produce more, faster, with less effort. Your output holds up under a quick glance. No one calls it shallow, because it looks polished. You may have no reason to think your thinking has gotten weaker, since you’re producing more than ever.
The cost comes due when you face a situation AI can’t handle – a new problem with no precedent, a high-stakes decision with incomplete information, a creative challenge that needs real originality. In that moment, you could discover the mental muscles you need are weaker than you assumed.
The Longevity Risk
The long-term risk isn’t that AI will replace your thinking. It’s that you’ll stop exercising the skills that do the hard work – and those skills may weaken like an unused muscle [2].
If the cost is real, it could stay invisible until you need the capacity. The first time you have to make a complex, high-stakes call without AI’s help – in a meeting, under pressure, with incomplete information – and find you can’t follow the chain of reasoning, that would be the cost coming due.
Looking ahead, we’d predict that the effects of AI use on judgment won’t be the same for everyone. That is a prediction, not an established finding. People who use AI as a scaffold for their own thinking – making the call themselves first, then comparing – may keep and even strengthen their judgment. People who use AI as a substitute – accepting output without checking it – may see a slow, unnoticed decline. The difference between those two paths isn’t the tool itself. It’s your relationship to the tool, and that’s something you get to choose.
The Protocol
The fix isn’t to reject AI. It’s to treat every interaction with AI as training for your own judgment.
Make the call yourself first. Then compare it with the AI’s answer. The gap between the two answers is where your growth happens. If the AI’s answer is better, figure out why. If your answer is better, that confirms your judgment is intact. Either result is useful. The only useless outcome is accepting the AI’s output without ever forming your own answer first.
This approach takes more time per task. That’s the point. The extra time isn’t wasted – it’s the mental training that keeps your judgment from weakening.
References
↑︎ Gerlich M. Societies. 2025;15(1):6. DOI: https://doi.org/10.3390/soc15010006
↑︎ Pascual-Leone A, Amedi A, Fregni F, Merabet LB. "The plastic human brain cortex." Annual Review of Neuroscience. 2005;28:377-401. DOI: https://doi.org/10.1146/annurev.neuro.27.070203.144216
Kurt Greiner
Kurt is a digital strategist and IT professional blending emerging technology with practical application to help businesses and individuals streamline their digital presence. His current work focuses on the intersection of intentional living and technological resilience, exploring how individuals can leverage modern tools to navigate the second half of life with purpose.
Leave a Reply