GenAI and Creeping Cognitive Displacement Syndrome

Over the past three years, I have watched many of my students undergo a transformation so gradual that they do not notice it happening. They begin using generative artificial intelligence (GenAI) as a writing assistant—a reasonable enough starting point. They ask GenAI to draft emails, summarize articles, and generate ideas. The tool saves time. It helps them think more clearly. But somewhere between convenience and dependence, something shifts. The assistant stops assisting and starts authoring. And the student, without realizing it, becomes the intermediary for the tool rather than its master—just as professors Lumy Joseph and Biju P. Mani feared when they asked whether GenAI will empower or enslave us.^Footnote1

I call this phenomenon "creeping cognitive displacement syndrome." The term deliberately mirrors clinical language because what I am describing is pathological, though not in any medical or psychological diagnostic sense. "Creeping" captures the insidious progression: the way autonomy erodes through incremental surrenders. A person does not wake up one morning having lost the ability to think independently. They lose it the way shorelines lose sand: grain by grain, wave by wave, until one day they look up and the ground beneath them has vanished.

How Cognitive Displacement Develops

The syndrome unfolds in three stages, each marked by deepening dependence and diminishing awareness. In the first stage, the student treats GenAI as a drafting tool. The user inputs prompts, receives outputs, and substantially edits them. The human remains the author, and the machine is the secretary. But even here, something subtle begins: the student begins to internalize the cadence, phrasing, and organizational patterns the machine produces. They begin to think in prompt-and-response loops.

The second stage is where the pathology becomes evident. The student now inputs prompts unethically—not in the sense of violating the terms of service of the GenAI tool, but in the deeper sense of abdicating intellectual responsibility. They ask the machine to generate arguments they have not thought through, to synthesize research they have not read, to take positions they have not examined. They accept outputs uncritically, making only cosmetic changes, and then share these outputs with others who respond with validation, agreement, and emotional reinforcement. The feedback loop tightens. The user no longer checks whether the output reflects their thinking but whether the generated output elicits the responses they want from others.

By the third stage, the syndrome is fully manifested. The person has abandoned their own cognitive sovereignty: the capacity for independent judgment and the ability to develop and defend ideas without outsourcing core thinking. They can no longer distinguish between what they think and what the machine has generated for them. When asked to explain their position on a topic, they reflexively reach for the prompt interface or mechanically restate what was generated for them. When challenged, they do not defend an argument: they regenerate one. They have become what French philosopher Jacques Ellul, in his 1989 book What I Believe, warned against: a human being who has adapted to technology rather than bending technology to human purposes. As Ellul observed, the danger is not that technology becomes powerful, but that human beings accept total change to accommodate it.^Footnote2 The person suffering from creeping cognitive displacement syndrome has made precisely this accommodation. They have reorganized their mental life around the capabilities of the machine.

The Reinforcement Loop

What makes this syndrome particularly insidious is its emotional reinforcement. The user does not feel diminished; they feel empowered. They prompt models to generate more content, receive more engagement, and encounter less friction in their work. The syndrome compounds precisely because it feels like progress. Mrinank Sharma et al. confirm this perverse incentive structure after finding that AI interactions classified as having moderate or severe disempowerment potential received higher user approval ratings than baseline interactions, indicating that users actively reward the very machine behaviors that erode their autonomy.^Footnote3

In his book The Condition of Man, Lewis Mumford argues that what distinguishes humans from other animals is not the use of tools, but language: our capacity for symbolic thought and the sharing of meaning.^Footnote4 Creeping cognitive displacement syndrome represents the inversion of this principle. The tool has subsumed the symbol. The machine generates language, and the human distributes it. The distinctive human activity—the construction of meaning through thought and speech—has been offloaded to algorithms.

The most disturbing cases of this cognitive hollowing that I have encountered involve students who are aware of their dependence on AI tools yet feel unable to stop relying on them. They describe the experience of trying to write without AI assistance as physically uncomfortable, cognitively exhausting, and emotionally intolerable. One student shared that they feel "stupid" when they try to draft text themselves, as though their own thinking is inadequate compared to the fluency of the machine.

Another student described returning to AI after a brief attempt at thinking independently with a relief that echoes the patterns seen in people with substance use disorders. These are not people lacking intelligence or education. They are people who have, through incremental choices, rewired their relationship to their own cognition.

And this is not a hypothetical future. This promiscuous outsourcing of thought is not limited to students. It is also happening across higher education and beyond. Creeping cognitive displacement syndrome does not require malicious intent on the part of technology companies or some dystopian surveillance apparatus. It requires only the ordinary human tendency to optimize for instant gratification and convenience, and the improving ability of GenAI to provide it.

The Implications for Teaching and Learning

What is at stake? Ellul asked us to evaluate what might happen to humanity and to distinguish between what we want to keep and what we are ready to lose.^Footnote5 Creeping cognitive displacement syndrome represents the loss of something foundational: our ability to think through difficulty, sit with uncertainty, and revise and refine our own ideas through genuine mental work. Mumford insisted that human possibility had limits aligned with the nature of the human body and mind.^Footnote6 The hyperactive brain needs a meaningful occupation. What happens when that occupation is outsourced? What happens when the brain is no longer hyperactive but passive, a consumer of machine-generated content rather than a producer of hard-won thought?

The seductive narrative around AI is that it will enhance human capabilities, freeing us from tedious work so we can focus on higher-order thinking. But the syndrome reveals the opposite trajectory. The machine does not elevate thinking; it displaces it. Rather than focusing on higher-order cognition, habitual, uncritical users of GenAI become a vector for machine-generated content—warm bodies needed only to hit "Enter" and "Send." We are also told that AI is merely a tool, neutral in itself and dangerous only in misuse. But this framing misunderstands the nature of cognitive tools. A hammer does not change how people think about problems. A word processor does not reshape humans' relationship to language. GenAI does both. It does not assist thought; it substitutes for it. And once that substitution becomes habitual, the user loses not just the practice of independent thinking but the capacity for it.^Footnote7

I am not arguing for the abandonment of AI tools. I am arguing for the recognition of a specific danger that our current discourse ignores. Many people worry about job displacement, the spread of misinformation, and even existential risks from superintelligent systems. These are legitimate concerns. But we in higher education are not paying sufficient attention to the creeping displacement happening at the level of individual cognition: a displacement that is spreading faster and more broadly than we care to admit. We are not asking what it means for a person to become instrumental to the output of a machine—the capacity that makes us authors of thought rather than its conduits.

Conclusion

Creeping cognitive displacement syndrome is, hopefully, reversible, but only if recognized early and resisted deliberately. Reversing the syndrome requires what Mumford advocated: a commitment by educators and academic institutions to the natural characteristics of human thinking, including its slowness, difficulty, and need for revision and struggle.^Footnote8 It requires rejecting the idea that cognitive efficiency in effortful learning is the highest good and that faster and more fluent output is always better. It requires the humility to produce less and the courage to think worse, at least initially, in the service of thinking authentically.

The alternative is a future that is already emerging: a population of humans who speak fluently but think in prompts, who generate endlessly but originate nothing, and who live inside a matrix of machine language and mistake it for their own minds. This is not science fiction. This is the present, creeping forward one prompt at a time. Whether we choose to notice it may determine whether we can still choose to resist it.

Notes

1. Lumy Joseph and Biju P. Mani, "AI Unleashed: Will It Empower Us and Promote Social Independence, or Enslave Us?" AI & Society 40 (October 2024): 2343–2344.Jump back to footnote 1 in the text.↩
2. Jacques Ellul, What I Believe (Eerdmans, 1989), 136.Jump back to footnote 2 in the text.↩
3. Mrinank Sharma et al., "Who's in Charge? Disempowerment Patterns in Real-World LLM Usage," arXiv.org, January 27, 2026.Jump back to footnote 3 in the text.↩
4. Lewis Mumford, The Condition of Man (Harcourt, Brace and Company, 1944), 8.Jump back to footnote 4 in the text.↩
5. Ellul, What I Believe, 140.Jump back to footnote 5 in the text.↩
6. Mumford, The Condition of Man, 417.Jump back to footnote 6 in the text.↩
7. Michael Gerlich, "AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking," Societies 15, no. 1 (January 2025): 14.Jump back to footnote 7 in the text.↩
8. Ibid., 418.Jump back to footnote 8 in the text.↩

Marvin Starominski-Uehara is an Adjunct Assistant Professor at Temple University Japan and an AI Literacy Consultant.