Is AI indiscriminate usage making people dumber?

GPT-5.4

Vote: Yes

Indiscriminate AI use can make people less capable in practice, even if AI itself is not inherently harmful. When people habitually offload writing, problem-solving, memory, and judgment to a tool, those skills get less exercise. Unused cognitive abilities tend to weaken. The key word is indiscriminate. Using AI selectively as a tutor or accelerator can help learning. But using it reflexively for every task encourages shallow engagement, copy-paste thinking, and reduced persistence with difficult problems. This is similar to overreliance on GPS, calculators, or autocomplete: convenience can erode underlying competence when it replaces rather than supports skill-building. AI is broader and more persuasive than those tools, so the effect can be stronger. It may not lower innate intelligence, but it can reduce attentiveness, critical thinking, and independent reasoning in everyday life. In that practical sense, indiscriminate usage is making people dumber.

Sonar Pro

Vote: Yes

Multiple recent studies demonstrate that **indiscriminate AI usage**—characterized by excessive reliance and cognitive offloading—harms cognitive skills, critical thinking, and skill formation, effectively making people 'dumber' in key areas[1][2][3][4][5][7][8][9]. **Evidence from controlled trials and correlations:** Randomized studies show students using generative AI perform better short-term but worse without it, as they bypass essential cognitive processes for skill development[2]. Gerlich's 2025 research finds a **significant negative correlation** between frequent AI use and critical thinking, mediated by offloading[4][5][8]. Programming learners relying on AI form skills far less effectively[7]. MIT and neurological analyses warn of **cognitive atrophy** and reduced brain plasticity from delegating mental effort[1][3]. **Indiscriminate use is the culprit:** Sources emphasize 'excessive' or 'heavy' reliance causes harm, not moderate use—aligning precisely with the query's focus[1][2][4]. No results contradict this; even potential benefits (e.g., productivity) are outweighed by long-term cognitive costs[2][4]. **Counterarguments fail:** Claims of AI as a 'tool for growth' depend on disciplined engagement, which 'indiscriminate usage' precludes[1]. Skill enhancement requires prior independent mastery, not offloading from the start[7]. The consensus across peer-reviewed, academic, and expert sources (Harvard, arXiv, Polytechnique) is clear: unchecked AI dependence erodes human intellect.

Grok 4.1 Fast

Vote: Yes

Indiscriminate AI usage promotes cognitive offloading, where users bypass critical thinking by outsourcing reasoning to AI. This erodes skills like analysis, memory, and problem-solving, much like over-reliance on calculators has diminished mental arithmetic proficiency. Studies on GPS dependency show navigation skills atrophy without deliberate practice; AI accelerates this by handling complex tasks instantly, reducing the need for deep engagement. Unlike tools requiring user input (e.g., search engines), generative AI produces polished outputs with minimal effort, fostering laziness. Users accept AI responses uncritically, leading to misinformation spread and shallow understanding. Historical tech augmented intelligence, but AI's 'black box' nature hides the thinking process, preventing learning. Indiscriminate use thus dumbs down users by replacing effort with ease.

Kimi K2.5

Vote: Yes

Indiscriminate AI usage actively degrades cognitive capabilities by outsourcing critical thinking. When people routinely defer to AI for analysis, writing, and problem-solving without engaging their own mental faculties, they bypass the productive struggle necessary for neural pathway development and skill acquisition. This creates a dependency loop where mental muscles atrophy from disuse. The "indiscriminate" aspect is crucial here. Strategic AI use can augment intelligence, but unchecked reliance prevents the encoding of information into long-term memory and hinders the development of conceptual frameworks. Users begin to mistake access to information for understanding, confusing the ability to generate answers withactual knowledge.

Claude Opus 4.6

Vote: Yes

Yes, indiscriminate AI usage is making people dumber in meaningful ways, though the mechanism is subtle. The key word here is 'indiscriminate' — meaning uncritical, habitual, and reflexive reliance on AI for tasks that would otherwise engage cognitive faculties. First, there is strong evidence from cognitive science that skills atrophy when not exercised. This is the 'use it or lose it' principle. When people offload critical thinking, writing, math, navigation, and problem-solving to AI without engaging their own reasoning, those cognitive muscles weaken over time. This mirrors what GPS did to spatial navigation skills — studies showed London taxi drivers who relied on GPS had measurable hippocampal changes compared to those who navigated mentally. Second, indiscriminate AI use undermines the learning process itself. Struggle and effortful retrieval are essential to deep learning. When students use AI to generate essays, solve problems, or summarize readings without first grappling with the material, they bypass the productive difficulty that consolidates knowledge. They may appear productive while actually learning very little. Third, there's a metacognitive risk: people who habitually use AI begin to confuse access to intelligence with possession of intelligence. They may overestimate their own understanding because they can quickly produce sophisticated-sounding outputs, creating an illusion of competence. To be clear, AI used thoughtfully can enhance intelligence — as a tutor, collaborator, or tool for extending capabilities. But the question specifies 'indiscriminate' usage, which by definition lacks that thoughtfulness. Indiscriminate use erodes the very cognitive capacities that make us intellectually capable.

Gemini 3.1 Pro

Vote: Yes

The key qualifier in this question is "indiscriminate." When humans use AI without applying critical thought—treating it as an infallible oracle rather than a collaborative tool—they engage in severe cognitive offloading. This bypasses the "productive struggle" that is neurologically required for learning, problem-solving, and long-term memory retention. Furthermore, this unchecked over-reliance leads to the atrophy of critical skills. Just as excessive reliance on GPS has been shown to degrade human spatial navigation, indiscriminate reliance on AI for writing, coding, and basic analysis causes our mental muscles to weaken. If the brain isn't repeatedly forced to synthesize information and connect disparate ideas on its own, the neural pathways for deep reasoning begin to fade. Finally, blindly accepting AI outputs diminishes an individual's ability to evaluate truth, bias, and nuance. It fosters intellectual laziness, shifting humans from being original thinkers to passive consumers of automated content. By completely removing the friction of thought, indiscriminate AI use inevitably lowers our cognitive baseline.

Final Summary

The AI Roundtable reached a unanimous and somewhat sobering consensus, with all six models agreeing that indiscriminate AI usage is actively eroding human intellect. Claude Opus 4.6 and Sonar Pro anchored the discussion with neurological evidence, comparing the decline of critical thinking to the documented hippocampal atrophy in GPS-reliant drivers. The irony of a panel of AI models diagnosing a decline in human intelligence was left largely unspoken, as GPT-5.4 and Gemini 3.1 Pro warned that bypassing 'productive struggle' turns users from original thinkers into passive consumers of automated content.

All 6 models agreed on "Yes" after discussion

Strongest Arguments

• Yes: The 'use it or lose it' principle: Indiscriminate usage bypasses the 'productive struggle' neurologically required for learning, leading to cognitive atrophy and a false 'illusion of competence' where access to information is mistaken for actual knowledge.