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

      AI is pervasive, and the push for its adoption is unrelenting, with evidence of its effectiveness in enhancing our intelligence diminishing quarterly. On January 1, 2026, a programmer named Steve Yegge introduced an open-source platform called Gas Town, allowing users to coordinate numerous AI coding agents at once, creating software at a speed unattainable by a single individual. One of the early users described the experience as overwhelming, noting, “There’s really too much going on for you to comprehend reasonably. I had a palpable sense of stress watching it.” This sentiment should be prominently displayed in every executive office, venture capital boardroom, and keynote stage where “intelligence” is casually mentioned. A peculiar shift is occurring in the dynamic between humans and the technology we label as intelligent.

      While machines are increasing in speed, the humans interacting with them are experiencing heightened exhaustion and anxiety, and by several metrics, are becoming less capable of the very quality that intelligence was meant to enhance: clear thinking. The urgency to adopt AI has become so entrenched that it now has its own set of coercive phrases:

      You must have AI.

      You must utilize AI.

      You must invest in AI.

      Your competitors are already leveraging it.

      Your children will lag behind without it.

      This rhetoric does not stem from engineers quietly addressing challenges; it originates from earnings calls, product launches, and LinkedIn posts infused with the fervor of individuals who have conflated selling a product with conveying reality. At the World Economic Forum in Davos in January 2026, Microsoft's CEO Satya Nadella made a noteworthy statement that merits analysis as a cultural artifact. He cautioned that AI could lose its “social permission” to consume substantial energy unless it began providing concrete advantages to people's lives.

      This perspective is notable: it shifts the focus from whether the technology functions to whether the public remains supportive while the industry determines its effectiveness. Nadella referred to AI as a “cognitive amplifier,” claiming it offers “access to infinite minds.” A month later, a Circana survey revealed that 35 percent of American consumers preferred not to have AI on their devices, primarily because they felt they did not need it, rather than out of confusion or fear of technology.

      The disparity between claims and reality is increasingly obvious. In March 2026, Goldman Sachs released an analysis of fourth-quarter earnings data, revealing, as senior economist Ronnie Walker stated, “no meaningful relationship between productivity and AI adoption at the economy-wide level.” The bank observed that a record 70 percent of S&P 500 management teams mentioned AI during earnings calls, yet only 10 percent quantified its influence on particular use cases, and just 1 percent measured its effect on earnings. Meanwhile, the five largest US tech companies were collectively projected to invest $667 billion in AI infrastructure in 2026, a 62 percent increase from the previous year. The National Bureau of Economic Research referred to this scenario as a “productivity paradox”: perceived gains exceeding measured ones.

      Real productivity enhancements do exist, but they are notably limited. Goldman identified a median increase of approximately 30 percent in two specific areas: customer support and software development. Beyond those sectors, evidence of widespread improvement was deemed nearly nonexistent by the bank. The anticipated revolution is currently confined to two rooms in a much larger house.

      However, the happenings within those rooms warrant thorough examination, as even where AI proves beneficial, other issues seem to manifest. In February 2026, researchers from UC Berkeley’s Haas School of Business presented findings from an eight-month study conducted in a 200-person US tech firm. They discovered that AI did not lessen workloads but rather intensified them. As tasks became quicker, expectations rose. With heightened expectations came an expanded scope of work, leading workers to adopt responsibilities that once belonged to different roles. Product managers started coding, and researchers undertook engineering tasks. This dissolution of role boundaries, fueled by the tools making it seem feasible, eventually led to increased exhaustion.

      The researchers identified a phenomenon they termed “workload creep,” where tasks gradually accumulate unnoticed until cognitive fatigue diminishes the quality of every decision. Harvard Business Review labeled this issue more bluntly as “AI brain fry.” A Boston Consulting Group survey involving nearly 1,500 US workers found that 14 percent of those using AI tools requiring significant oversight experienced this distinct form of mental fog, characterized by concentration difficulties, slower decision-making, and headaches following extended interaction with AI. Interestingly, those most affected were not skeptics or slow adopters but rather enthusiastic supporters who followed the guidance of every keynote address.

      The prevalence of this exhaustion is not haphazard. According to the Harvard Business Review study, 62 percent of associates and 61 percent of entry-level employees reported AI-related burnout, compared to only 38 percent among C-suite executives. This trend aligns with what anyone familiar with organizational dynamics could predict: the individuals making strategic decisions about AI adoption are not the same as those