Exasperated...
In the year 2025:
- “Wait—there’s no GPS signal? Oh my god, I’m completely lost. I don’t even remember how to read a map!”
- “I was in Milan and the translation app froze. I just stood there—I couldn’t ask how much the train ticket cost. I use the app for everything. I never bothered to remember the words!”
In the year 2055:
- “The AI was down and I had to email the whole team about the new policy. I haven’t written anything from scratch in years. I didn’t even know where to start!”
- “She said she couldn’t make a decision until she checked the model’s recommendation. Not wouldn’t—couldn’t!”
In the early 2000s, the rapid advancement of technology ushered us from the age of analog memory into the era of convenience. Calculators made mental arithmetic optional. GPS replaced spatial reasoning with turn-by-turn obedience. Search engines offered instant recall, making memory feel obsolete. These tools made our lives easier, more efficient, more comfortable. And they changed our habits—altering where we invest time, attention, and effort.
Then came AI. And with it, a new kind of cognitive convenience: more seamless, more seductive, more insidious. It increasingly anticipates, automates, and substitutes for particular cognitive acts—summarizing, drafting, recommending. What we once rehearsed ourselves, we now delegate. And practice withers. We like the readiness, the availability, the frictionless fluency. We get used to it, and we become increasingly dependent on it.
So when the moment demands unscripted clarity and instantaneous response—when the GPS fails, the client asks a follow-up, or the conversation veers off-script—many struggle. This is not a failure of intelligence; it is a failure of readiness. And readiness, unlike information, cannot be downloaded.
We are not just changing how we think. We are changing our capacity to prepare, to respond, to act without assistance. The more fluent the technology becomes, the more we defer to it—not out of necessity, but out of habit. And habit, once formed, reshapes capability.
This is not a critique of the technology—we are not asking whether AI and the convenience it brings are good or bad, for it is impartial. It is also not a rejection of new norms, nor a lament for a bygone era. For we have seen how AI, when used with intentionality and care, can bring genuine cognitive and emotional benefits—such as in the case of certain older adults, where it offers a tailored form of companionship and mental stimulation, or in other instances where it provides the patient, judgment-free support that helps foster engagement and growth for some neurodivergent individuals.
The concern here is dependency and overreliance. As we turn to it for every answer, every decision, every moment of uncertainty—not because we must, but because we no longer practice the alternative—are we letting our mental acuity, and our very capacity to reason, reflect, retain, and respond to complexity, become eroded? How do we even resist a drift so seamless, so convenient that it feels almost inevitable?
Illuminating Arches on Modern Preparedness
We turn to the early modern philosophers. Their world was shifting—religious authority waning, scientific reasoning emerging, and inherited structures of thought being questioned. It was a time of cognitive reorientation, when the foundations of knowledge were no longer fixed, and the act of thinking itself became a subject of inquiry. Much like today.
- Michel de Montaigne, the introspective essayist, believed that understanding grows through reflection, not accumulation. He saw thinking as a slow, personal discipline—one that resists shortcuts and thrives in ambiguity. His essays were a deliberate and meditative journey to show that wisdom is found not in final answers, but in the honest pursuit of the question.
- Blaise Pascal, the mathematician-philosopher, viewed thought as a confrontation with the human condition—a way to prepare for truth, decision, and mortality. He argued that humanity often distracts itself with entertainment and trivial pursuits to avoid confronting its own fragility and the great questions of existence. For him, comfort without clarity was a dangerous indulgence.
- Baruch Spinoza, the rational ethicist, held that reason is not just a tool. It is a path to freedom. He believed that clarity must be earned, and that understanding is inseparable from ethical living. He sought to liberate the mind from passion and superstition by building a rigorous, geometric system of thought.
When Convenience Erodes Cognitive Rigor
Cognitive drift—the erosion of mental preparedness in the face of AI convenience—is not hypothetical. It is a phenomenon that is already troubling educators, researchers, ethicists, workplace designers, and policymakers worldwide. A growing body of studies points to the costs of over-reliance: diminished reasoning, weaker retention, and a steady decline in critical engagement.
A systematic review in Smart Learning Environments (Zhai et al., 2024) found that students using AI dialogue systems often adopted “efficiency heuristics”—shortcuts that favored fast, polished answers over sustained reasoning. This over-reliance was found to impact critical cognitive abilities, including decision-making and analytical thinking. A field experiment led by researchers at the University of Pennsylvania, titled Generative AI Without Guardrails Can Harm Learning (Bastani et al., 2025), found a striking pattern: students using a standard GPT tutor performed ∼48% better during practice, yet scored ∼17% lower on the final exam than peers who studied without AI. AI use appeared to have interfered with deeper independent cognitive processing and long-term retention.
Similar patterns are being observed elsewhere. A case study from Ho Chi Minh City University of Technology and Education (Nguyen et al., 2024) found that students readily used ChatGPT for idea searching, assignment completion, and information retrieval, a reliance that, as other contemporaneous studies from the region confirm, has led to concerns about diminished creativity and independent thinking. A perspective piece in Frontiers in Psychology (2025) described this as a “cognitive paradox”: AI tools can boost short-term performance and efficiency but risk eroding higher-order thinking skills—analysis, evaluation, and synthesis—if used uncritically.
Policy institutions are beginning to respond. For instance, the Council of Europe has warned that AI in education must support, not supplant, human agency, and must be governed by human-rights principles. The European Commission’s Joint Research Centre has likewise emphasized the cognitive and ethical risks of AI-enabled learning, urging safeguards that preserve critical thinking, intellectual autonomy, and human oversight.
Cultivating Cognitive Strength in the Face of Drift
If Montaigne, Pascal, and Spinoza were here today, what might they advise to mitigate against this cognitive drift?
Montaigne: Invite Friction into the Act of Knowing
Montaigne did not believe in passive absorption. To him, the mind was not a vessel to be filled but a muscle to be exercised. He would likely urge us to resist the seduction of seamlessness and instead let the process of learning be marked by pause, doubt, and the slow churn of thought.
In some university classrooms in the U.S. and Europe, instructors are now asking students to annotate AI-generated responses using platforms like Hypothesis. The task is not correction—it is interrogation. Students must explain why a claim holds, where it falters, and how it might mislead. This is not mere critique; it is a rehearsal of judgment.
Beyond the classroom, reflective journaling platforms such as Reflection.app and research-driven projects such as MindScape use AI not as an oracle but as a mirror, guiding users toward self-awareness, emotional clarity, and personal coherence—emphasizing depth rather than speed. A rhythm that Montaigne himself would probably approve.
Pascal: Design for Discomfort
Pascal understood the human impulse to flee from silence. Distraction, he argued, is more than mere noise—it is a kind of spiritual anesthesia, a way of avoiding the confrontation of thought, the weight of mortality, and the demands of meaningful judgment. Today, as AI offers instant answers and curated calm, he might ask: How do we preserve the mental rigor real-world decisions require? Where is the space for unease?
To answer that, one can look to practices that deliberately invite disquiet, training the mind to resist cognitive drift. At the University of Melbourne, interactive oral assessments place students in unscripted, real-time dialogue, often simulating professional scenarios. There is no predictive text, no second draft—only presence. This discomfort is intentional: it trains the mind to think under uncertainty.
At the University of Saskatchewan, educators are designing AI-resistant tasks such as live debates, role plays, and branching case simulations. By demanding spontaneity and judgment, these exercises preserve faculties that cannot be outsourced to machines.
These examples illustrate a single principle: deliberate, uncomfortable engagement—with ideas, with others, with oneself—cultivates the mental preparedness real-world decisions demand. Pascal would likely recognize in these efforts a return to the necessary exigence of thinking.
Spinoza: Anchor Intelligence in Ethical Clarity
Spinoza did not fear intuition, but its detachment from understanding. For him, ethical action was never a matter of compliance or instinct. It was the fruit of coherence—the slow, deliberate alignment of thought, emotion, and consequence. To resist cognitive drift, Spinoza would insist, we must do more than slow down; we must clarify. We must ask not only what AI can do, but also what we are becoming as we allow it to act on our behalf.
This principle finds expression in practice. At Taipei Veterans General Hospital, clinicians use AI-assisted diagnostic tools not as final arbiters, but as provocateurs of discussion: multidisciplinary teams review AI-generated recommendations to ensure they align with patient history, clinical nuance, and ethical considerations. At Radboud University Medical Center in the Netherlands, radiologists engage in a “human-in-the-loop” protocol, scrutinizing the algorithm’s assumptions, identifying potential bias or outdated data, and anticipating deviations from current clinical standards. In these cases, the deliberate and reflective process intentionally resists automation, focusing instead on ethical clarity and coherence.
In the corporate world, simulation-based ethics training confronts the same drift. Programs such as Where Do You Draw the Line? and E.I. Games’ Business Ethics Simulation immerse participants in real-time dilemmas—conflicts of interest, breaches of confidentiality, pressures to compromise. There are no algorithmic shortcuts. Participants must navigate ambiguity, defend their reasoning, and confront the consequences of their choices. These exercises do not promise efficiency; they aim to cultivate preparedness.
The risk of cognitive drift is acute. For AI has simply made it too easy, too available, too frictionless.
This architecture of mind—the capacity to reason, discern, remember, and make sense—is ours to lose. And the cost is real: not overt, but persistent, pervasive, and systemic. Unless we choose otherwise—and act. Deliberately. Starting now.
From the AI Conundrums and Curiosities: A Casual Philosophy Series by Jacquie T.
Comments ()