AI is enhancing our speed and productivity but also diminishing our critical thinking skills.

      AI is ubiquitous, the demand for its adoption is ceaseless, and the evidence supporting its claim to enhance our intelligence is diminishing quarterly.

      On New Year’s Day 2026, a programmer named Steve Yegge introduced an open-source platform called Gas Town, which enables users to coordinate multiple AI coding agents at once, creating software at a pace unattainable by any individual.

      One of the initial users described the experience as overwhelming. "There’s really too much going on for you to comprehend reasonably," he stated. "I had a palpable sense of stress watching it."

      That observation should be prominently displayed in every executive office, venture capital meeting room, and at every CES keynote where the term “intelligence” is casually thrown around. An unusual dynamic is developing between humans and the technology we label as intelligent.

      While machines are becoming faster, the humans that engage with them are increasingly fatigued, anxious, and, according to several indicators, less able to do the very thing that intelligence was meant to improve: clear thinking.

      The push to adopt AI has become so intense that it has established its own coercive lexicon:

      You must have AI.

      You need to implement AI.

      You need to purchase AI.

      Your competitors are already employing it.

      Your children will lag without it.

      This rhetoric does not originate from engineers quietly tackling issues. Instead, it arises from earnings calls, product launches, and LinkedIn updates churned out with frantic energy from those who conflate marketing a product with representing the truth.

      At the World Economic Forum in Davos in January 2026, Microsoft CEO Satya Nadella articulated a phrase so insightful it warrants examination as a cultural artifact. He cautioned that AI risked losing its "social permission" to use significant amounts of energy unless it began to provide verifiable advantages to people's lives.

      The perspective was notable: it was not about whether the technology functioned, but whether the public could be persuaded to stay supportive while the industry determined its effectiveness. Nadella referred to AI as a "cognitive amplifier," providing "access to infinite minds."

      A month later, a Circana survey indicated that 35% of US consumers were opposed to having AI on their devices. The main reason wasn’t fear or confusion; it was simpler: they felt they didn’t need it.

      The disparity between the hype and actual evidence is becoming increasingly apparent. In March 2026, Goldman Sachs released an analysis of fourth-quarter earnings showing, in the words of senior economist Ronnie Walker, “no meaningful relationship between productivity and AI adoption at the economy-wide level.”

      The bank noted that 70% of S&P 500 management teams mentioned AI during their earnings calls, but only 10% had quantified its impact on specific use cases, and just 1% had measured its effect on earnings. Meanwhile, the five largest US tech companies were anticipated to spend $667 billion on AI infrastructure in 2026, a 62% rise from the previous year.

      The National Bureau of Economic Research characterized this scenario as a "productivity paradox": perceived gains outstripping measured ones.

      There are indeed genuine productivity advancements, but they are remarkably limited. Goldman found a median enhancement of around 30% in two particular areas: customer support and software development. Beyond those fields, the overall evidence for significant improvement was, according to the bank, virtually nonexistent. The anticipated transformation, for the moment, is restricted to a few rooms in a very large building.

      However, what is happening in those rooms merits close scrutiny, as even where AI succeeds, something else seems to be going awry.

      In February 2026, researchers at UC Berkeley’s Haas School of Business unveiled findings from an eight-month study conducted with a 200-person US tech firm, revealing that AI did not lessen workloads but instead intensified them. Tasks were completed more rapidly, leading to heightened expectations. As expectations grew, so did the task scope. The involvement expanded, prompting workers to assume responsibilities previously designated to other roles. Product managers began coding, and researchers took on engineering duties. Role boundaries blurred due to the tools’ perceived capabilities, resulting in widespread exhaustion.

      The researchers identified a cycle they termed "workload creep": a gradual accumulation of tasks that goes unnoticed until cognitive fatigue erodes the quality of every decision.

      Harvard Business Review referred to this phenomenon with a more straightforward term: "AI brain fry." A Boston Consulting Group study involving nearly 1,500 US workers indicated that 14% of those using AI tools that needed considerable oversight experienced this distinct form of mental fog, marked by difficulty focusing, slower decision-making, and headaches following prolonged AI use.

      The workers most impacted weren’t the skeptics or the slow adopters. They were the zealous early adopters, those who followed every keynote directive.

      The distribution of this exhaustion is not random. According to the Harvard