iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
AI can make it easier to avoid some of the thinking a task requires—but the evidence does not show that ordinary AI use inevitably reduces intelligence or causes lasting cognitive decline. A controlled student study found weaker reasoning on one short research task when students used ChatGPT 3.5 instead of Google; a workplace survey found that users often perceived less cognitive effort when using generative AI. The practical risk is accepting a fluent answer without checking, explaining, or integrating it yourself.
What the evidence says about AI and thinking
The strongest direct evidence in the cited studies is specific, not universal. In a 2024 randomized experiment, 91 university students researched nanoparticles in sunscreen using either ChatGPT 3.5 or Google. The ChatGPT group reported lower cognitive load and produced recommendations with lower-quality reasoning and argumentation. The groups did not differ significantly in perspective diversity. This was one brief task, one model version, and a university population—not a test of long-term learning or all possible ways to use AI. Read the study by Stadler, Bannert, and Sailer.
A separate 2025 CHI study surveyed 319 knowledge workers who provided 936 examples of generative AI use at work. Respondents described lower effort in several cognitive activities. In the examples, 72% reported “less” or “much less” effort for Knowledge, 79% for Comprehension, 69% for Application, 72% for Analysis, 76% for Synthesis, and 55% for Evaluation. Those percentages describe reported effort categories in the study dataset; they are not measured drops in ability. The survey records perceptions and experiences, so it cannot show that AI caused lasting deterioration in critical thinking. Read the CHI 2025 study by Lee and colleagues.
Less effort is not the same as less ability
An easier task can mean the tool has removed routine work while leaving the important judgment to a person. It can also mean the tool has taken over part of the reasoning, or that a user interprets lower general effort as doing less critical thinking. Effort, engagement, and durable ability are related but distinct; neither a survey response nor a single task measures long-term skill by itself.
The student experiment matters because it measured the quality of recommendations as well as perceived cognitive load. In that task, reduced load coincided with weaker reasoning and argumentation. That is a reason to pay attention when the goal is learning or deep inquiry—not proof that every AI-assisted task produces the same outcome.
#1 Best Overall
- A good option for a Book Lover
- It comes with proper packaging
- Ideal for Gifting
Where the thinking goes when AI helps
Generative AI often shifts the work rather than eliminating it. The workplace study describes three changes that help explain why a user may feel less effort while still having important responsibilities:
From finding information to verifying it
An AI tool can assemble an answer quickly, but a person still needs to check factual claims against reliable sources and relevant expertise. A polished response is not evidence that its claims are accurate.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problemsFrom solving the problem to integrating a response
Generated material must fit the actual question, audience, and constraints. The user has to decide what applies, what is missing, and what should be rejected or revised.
From executing the task to stewarding the system
When AI performs part of the work, a person may spend more time directing and monitoring it. The human remains responsible for the result, especially where errors have consequences.
When AI use is more likely to help—or to short-circuit learning
The key question is whether the tool removes low-value friction or takes over the thinking you need to practice. AI is more likely to support the work when you can still explain why the result is appropriate, verify consequential claims, and make the final decisions. It is more likely to short-circuit learning when the goal is to build a skill but you accept the answer before attempting the reasoning yourself.
- For routine work: using AI to reduce clerical effort may be useful if you review the output and remain accountable for it.
- For learning: asking for a finished answer first can bypass the attempt, explanation, and correction that make practice valuable.
- For consequential decisions: a plausible answer should be treated as a starting point for verification, not as authority.
How to use AI without handing over the important thinking
For study and skill practice
- Make an initial attempt before asking AI for a solution. Write down your answer or the reasoning you used.
- Ask the tool to critique your explanation, identify assumptions, or point out a gap rather than simply supplying a replacement answer.
- Check corrections against course materials or other reliable sources, then revise the explanation in your own words.
- Before moving on, make sure you can explain the reasoning without relying on the generated response.
This sequence is a practical recommendation based on the studies’ emphasis on verification, integration, and foundational problem-solving; the cited studies did not test it as a formal intervention.
For work and decision-making
- Verify important factual claims against reliable external sources.
- Check that the output fits the actual task, audience, and constraints instead of accepting a generic response.
- Keep a person responsible for the final judgment and result, and monitor the system when it performs part of the work.
What remains uncertain
The student experiment involved a German university population, ChatGPT 3.5, and a short research-and-recommendation task. It does not establish how current models affect other ages, subjects, or longer-term learning. The workplace survey recruited people who used generative AI at work at least weekly and relied on self-reported examples; its authors note that respondents may conflate reduced general effort with less critical thinking. The study population also skewed younger and more technologically skilled, and the authors call for task-based and longitudinal research.
Lee and colleagues summarize the concern cautiously: “Moreover, while GenAI can improve worker efficiency, it can inhibit critical engagement with work and can potentially lead to long-term overreliance on the tool and diminished skill for independent problem-solving.” That is a stated possibility, not a finding that the study measured long-term harm.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
