Are We Getting
Stupider?
Cognitive atrophy, the GPS effect and what happens to human intelligence when thinking becomes a service you can subscribe to.
In 2011, a team at Columbia University ran a simple experiment. They gave participants a set of trivia questions and, before answering, told some of them that the answers would be saved to a computer — and others that they would be erased. The group who thought the answers would be saved remembered significantly less. They hadn't bothered to encode the information into memory because they knew it was stored elsewhere.
The researchers called this the Google Effect. When we know an external system holds information, we stop bothering to hold it ourselves. The brain, ever efficient, deprioritises encoding what it believes it can retrieve on demand.
That was 2011. Google only stored information. It did not think.
What happens when we outsource not just the storage of information, but the process of reasoning about it?
The GPS Brain
The most rigorously studied example of technology-induced cognitive atrophy is not smartphones or social media. It is GPS navigation — and what it did to the hippocampus.
London taxi drivers, famously, must memorise the city before receiving their licence — a process called “the Knowledge” that can take years. Neuroimaging research found differences in their hippocampi compared with drivers who follow fixed routes. The hippocampus is central to spatial memory and navigation. The work does not reduce the brain to a muscle with one simple rule; it does show that demanding navigation and long practice leave a trace. The brain responds to demand.
GPS changes the terms of the task. When the device carries the route, the user has less reason to build an internal map. Research on navigation supports that concern: delegating wayfinding can leave people with poorer route and survey knowledge. The long-term neurological consequences are still being investigated, and they should not be compressed into a single dramatic coefficient. But the direction is familiar. What is no longer practised is less available when the screen goes dark.
This is not merely a metaphor. Practice changes performance; sustained experience can shape the systems that support it. Use it or lose it is too blunt to be a complete theory of the brain. It is, however, a useful warning about skills we quietly stop exercising.
Now consider what AI is doing — and how much more of the cognitive stack it touches than GPS navigation.
AI Apathy — The New Cognitive Risk
The label is less important than the mechanism: cognitive offloading. We move a task from the mind to an external system, gain speed, and risk losing the practice that made us competent without it. With AI, the delegated work is no longer only recall or navigation. It can be the first interpretation, the first argument, the first diagnosis, the first draft of a decision.
Early research on generative AI does not yet justify a universal verdict on human intelligence. It does point to a real split. Used with structure — critique, verification, explanation, revision — AI can support reasoning and learning. Used as frictionless substitution, it can encourage shallow engagement and reduced epistemic effort. The mind reaches for the answer before it has attempted the problem.
The Pilot Problem
Aviation saw the problem early. Modern aircraft use automation to make flying safer, more consistent and more manageable. Yet aviation authorities still treat automation reliance as a training issue: pilots need proficiency both with the system and without it.
When automated systems fail, or behave in a way the crew does not expect, the human has to understand the situation and act. The response is not nostalgia for manual flight. It is deliberate practice: recurrent training, scenario work and a continuing ability to take control. The automation stays. The skill is maintained against the pull of convenience.
The parallel to AI-assisted knowledge work is direct. A developer who accepts a proposed patch without tracing its assumptions has not reviewed it. A researcher who accepts a summary without opening its sources has not learned from it. The danger is not that the tool offers help. It is that a plausible output can imitate understanding well enough to make its absence invisible.
The Google Effect, Supercharged
The original Google Effect altered how we encode information. AI alters something deeper: how we process it.
Google told you where to find the answer. You still had to read the page, evaluate the source, synthesise across multiple results, and form a judgement. The cognitive work of processing information remained yours. You just outsourced the retrieval.
AI can perform much of that processing. It can read, synthesize and propose a conclusion in one motion. When the user checks the trail, challenges the framing and supplies the judgment, the tool can extend thought. When the user merely accepts the conclusion, the role collapses from thinker to prompt author and output accepter.
The research is not unambiguous, because the tool and the contexts are new. The mechanism is clear enough to take seriously: offloading reduces the effort invested in a task, and skill depends on practice. The open questions are which abilities are affected, under what conditions, how fast, and how reversible.
What Atrophies — and What Doesn't
The research does not support a simple "AI makes you dumb" narrative. The picture is more specific — and in some ways more concerning.
Surface creativity — novel combinations, aesthetic experimentation
Information retrieval speed — finding facts, summarising material
First-draft production — getting something on the page quickly
Independent reasoning — reaching conclusions without external scaffolding
Deep memory encoding — retaining and connecting knowledge over time
Epistemic scepticism — the habit of questioning claims and demanding evidence
The pattern is consistent: AI handles the effortful parts of cognition — the parts that are uncomfortable, slow, and resist automation. Those effortful parts are also, not coincidentally, the parts that build intellectual capacity when practised. The comfortable, fast, generative parts remain with the human. The hard parts — the parts that make you better at thinking — are outsourced.
The Divergence Hypothesis
The most troubling scenario is not uniform cognitive decline. It is divergence.
A small group of people will use AI as a genuine cognitive amplifier — as a tool that frees mental bandwidth for higher-order thinking, that challenges their assumptions, that accelerates genuine learning. These people will become more capable, faster, with access to more breadth than any previous generation could match.
A much larger group will use AI as a cognitive outsourcing service — delegating thinking to it, accepting outputs without scrutiny, gradually losing the habit and the ability to reason independently. For this group, AI will deliver the appearance of competence while quietly eroding its foundations.
The danger is not that AI makes everyone stupider. It is that it makes the divergence in human cognitive capacity wider than it has ever been — and does so invisibly, through a mechanism that feels, at every individual moment, like efficiency.
The Irreversibility Question
Use-dependent plasticity is not a one-way door. Skills can be recovered through deliberate practice. The difficult part is noticing the loss early enough to care about rebuilding it.
But recovery requires awareness of the loss, and motivation to rebuild — two things that are hard to sustain when the AI is always there, always faster, always producing something that looks like a good answer. The incentive structure pulls toward deeper dependency, not toward the uncomfortable work of rebuilding what has been lost.
Aviation answers the problem with recurrent training: keep the human in practice even when the automation is excellent. Education is beginning to test similar boundaries through AI-restricted assessment. Most knowledge work has no equivalent. There is no mandated hour in which the analyst, lawyer, founder or engineer must work unaided and discover what they no longer know.
The question “are we getting stupider?” does not yet have a single, clean answer. But it has a serious one: where people routinely delegate a cognitive function without checking, practising or rebuilding it, capability is at risk. AI makes that delegation easier, broader and more seductive than the tools that came before.
Whether that matters depends entirely on whether those domains remain important in a world where AI can perform them. That is a question the next thirty years will answer. But the atrophy begins now, in a decision made every time we reach for the tool instead of attempting the thought.