We are handing the most powerful tool in human history to a population that is increasingly unprepared to question what it tells them. That is not a technology problem. It is a human one.
Artificial intelligence is accelerating at a pace that no policy, curriculum, or regulation has kept up with. At the same time, the foundational human skills needed to use it wisely, critical thinking, curiosity, source verification, and the ability to distinguish a summary from an interpretation, are measurably declining.
These two trends are not separate problems. They are a collision course.
This is not about being anti-AI. It is about understanding what is actually at stake when a tool this powerful is placed in unprepared hands.
The Usage Gap Is Already Here
AI adoption among students has moved faster than almost any technology in history. According to research compiled by DemandSage (January 2026), overall AI usage among university students jumped from 66% in 2024 to 92% in 2025. Generative AI usage specifically for academic assessments rose from 53% to 88% in the same period.
Source: https://www.demandsage.com/ai-in-education-statistics/
Meanwhile, a 2024 global survey by the Digital Education Council found that only 5% of students reported being fully aware of their institution’s AI guidelines, and 80% said AI in universities was not meeting their expectations.
Source: https://www.digitaleducationcouncil.com/post/what-students-want-key-results-from-dec-global-ai-student-survey-2024
Students are using AI constantly. Almost none of them have been taught how to use it well. That gap is not small. It is the size of the entire conversation we are not having.
Literacy Was Already in Trouble Before AI Arrived
The AI literacy crisis is landing on top of a general literacy crisis that was already underway. According to the National Center for Education Statistics (NCES), the percentage of U.S. adults aged 16 to 65 who fall in the lowest literacy levels rose from 19% in 2017 to 28% in 2023.
Source: https://sjhexpress.com/opinion/2025/02/28/the-decline-of-literacy-and-the-rise-of-ai-are-we-losing-the-ability-to-think/
54% of U.S. adults currently read below the equivalent of a sixth-grade level.
Source: https://www.thenationalliteracyinstitute.com/2024-2025-literacy-statistics
The OECD’s 2022 PISA assessment recorded the most significant decline in reading and mathematics scores since the assessment began, with many national systems losing the equivalent of three-quarters of an academic year in reading comprehension.
The 2024 Survey of Adult Skills confirmed the trend is continuing.
Source: https://jamescosullivan.substack.com/p/critical-thinking-was-in-decline-before-ai
We were already behind. AI did not create this problem. But it has made it urgent in a way that can no longer be ignored.
Critical Thinking Is in High Demand and Short Supply
Here is the central paradox: the very skill that is declining most sharply is the one the modern economy most desperately wants. According to the World Economic Forum’s
Future of Jobs report (2023-2027), analytical thinking, a direct synonym for critical thinking, is the most sought-after skill across industries worldwide. In a survey of 803 companies employing 11.3 million workers, critical thinking was rated more essential than technological literacy, AI proficiency, leadership, and even cybersecurity.
Source: https://pursuit.unimelb.edu.au/articles/our-kids-are-missing-out-on-critical-thinking
Business leaders know this. 92% of marketers surveyed in 2024 said they believe AI literacy will be a must-have skill within 2 to 4 years, and 69% of organizational leaders now identify AI literacy as essential to daily workflows, up 7% from the prior year.
Source: https://www.certlibrary.com/blog/introducing-the-2025-state-of-data-and-ai-literacy-report/
The demand is rising. The supply of people who actually think this way is not keeping pace.
Curiosity Is the Skill Nobody Can Teach Easily — And We Are Losing It
Underneath every good prompt, every smart follow-up question, every refusal to accept a summary at face value, is one thing: curiosity. The genuine desire to know more, to verify, to push deeper rather than accept the first answer that sounds authoritative.
Curiosity cannot be installed like a software update. It has to be cultivated from early childhood through an environment that rewards questions more than it rewards correct answers. And the evidence suggests that environment is disappearing.
A study by the MIT Media Lab, reported in Time magazine, found that students who used AI to write their essays showed the lowest brain activity of any group studied, because they bypassed the effortful, questioning process through which real understanding is built. The struggle is not a bug. It is where the learning happens. Remove the struggle, and you remove the growth.
Source: https://africasacountry.com/2025/09/the-demise-of-curiosity
European youth now consume over 34GB of media daily. Researchers have begun calling the cognitive result “clip thinking,” the ability to rapidly scan and discard information without ever pausing to interrogate its source, its accuracy, or its implications. It is not stupidity. It is adaptation to an environment that no longer rewards depth.
Source: https://www.20something.be/critical-thinking-young-people/
Instant answers delivered before the question has fully formed do not build understanding. They replace it. And once the habit of passive consumption is established, it is very hard to reverse.
AI Is Only as Conscious as the Person Asking the Questions
This is the point that almost never makes it into mainstream AI coverage: the quality of what AI gives you is almost entirely determined by the quality of what you bring to it. Prompting AI well is not a technical skill. It requires knowing how to ask precise questions, how to distinguish an interpretation from a fact, how to push back on a confident-sounding answer, and how to verify before trusting.
These are the same skills that have been in decline for a decade. Which means the people most likely to use AI as a passive confirmation machine, absorbing whatever it says without question, are the same people who already struggle to evaluate information critically.
A 2025 report by New America found that current AI literacy efforts are overwhelmingly superficial, focused on tool usage rather than the underlying thinking skills required for responsible engagement. Students are being taught to use the tool, not to think with it.
Source: https://www.newamerica.org/education-policy/briefs/foundational-skills-digital-literacy-in-the-age-of-ai-analysis-and-voices-from-the-field/
Meanwhile, the OECD estimates that AI could now outperform 90% of adults on standard literacy tests. Even the highest-performing countries cannot supply more than a quarter of their workforce with skills sufficient to outperform AI on these benchmarks.
Source: https://www.oecd.org/en/publications/is-education-losing-the-race-with-technology_73105f99-en.html
The bottleneck is not the technology. It is the quality of the human asking the questions.
What Needs to Change
We need to stop treating AI literacy as a technical elective and start treating it as a foundational survival skill, no different from reading or mathematics. Not “here is how to use ChatGPT,” but rather: here is how to think when you use any AI. Here is how to verify. Here is how to tell the difference between a summary and an interpretation. Here is why your next question matters more than the first answer.
This means starting earlier. It means training teachers before students. It means parents understanding what their children are encountering before they encounter it. And it means policymakers treating this with the same urgency they would any other public health issue, because an uncritical population armed with AI is, in every meaningful sense, a public health issue.
Digital Promise, in their November 2025 AI Literacy Imperative briefs, put it plainly: AI literacy must be built as a foundational skill across all disciplines and all audiences, not as a niche technical curriculum.
Source: https://digitalpromise.org/2025/11/10/the-ai-literacy-imperative-new-briefs-to-guide-ai-literacy-implementation-across-learning-environments/
The technology is not waiting for us to catch up. The question is whether we decide that curiosity, critical thinking, and the willingness to question are worth protecting, or whether we hand those responsibilities over to a machine and hope for the best.
Based on the evidence, hoping for the best is not a plan.
Sources referenced in this post are linked inline. This is part of an ongoing series on AI literacy, human cognition, and what it means to think well in an AI-saturated world.
0 Comments