AI, Attention, and the Mind

Your attention is one of the most valuable things you possess. It is the lens through which all of your experience passes. What you attend to shapes what you learn, what you feel, who you become. For most of human history, the primary competition for your attention was the physical world around you and the demands of the people in it. Today, an additional competitor exists: AI systems specifically engineered to capture and retain your focus. Understanding how these systems work, what they do to the mind that engages with them, and what you can do about it is not optional knowledge for people living in the 21st century. It is essential.

Recommendation Algorithms: An Attention Economy

The major social media and video platforms — TikTok, YouTube, Instagram, Twitter/X, Spotify — are powered by recommendation algorithms that decide what you see next. These algorithms are trained on billions of data points about user behavior: what you watched, how long you watched, what you paused on, what you scrolled past, what you searched for. Their objective function is straightforward: maximize engagement, which operationally means time on platform. AI researcher Tristan Harris, a former Google design ethicist, has described these systems as a race to the bottom of the brainstem — a competition to discover which content patterns trigger the deepest and most durable human psychological responses. The winner is not the most educational or the most beautiful content. It is the most emotionally compelling content, which often means the most outrage-inducing, the most validating, or the most novelty-producing. These systems are extraordinarily powerful and extraordinarily optimized — not for your benefit, but for your engagement. This is the attention economy: a system in which your attention is the commodity, sold to advertisers, and AI is the primary tool for extracting it.

Research on the cognitive effects of heavy social media use — much of it still contested and evolving — suggests several patterns worth knowing. First, frequent context-switching between short pieces of content may reduce the capacity for sustained attention. The brain adapts to its regular demands; if you regularly ask it to attend to five-second clips, it may become less comfortable with the sustained focus required by a long book, a long conversation, or a long problem. Second, algorithmic content selection may reduce the diversity of perspectives you encounter. If the algorithm learns that you engage more with content that confirms your existing views, it will serve you more of it — not out of malice, but because that pattern maximizes engagement. The result is a filter bubble: a personalized information environment that becomes progressively less diverse and more self-confirming. Third, there is evidence that the mere presence of a smartphone — even face-down, even silenced — reduces available cognitive capacity for the task in front of you, because part of your mind is monitoring for potential notifications. The device has become an attention competitor even when it is not being used.

Cognitive Offloading and What We Lose

Beyond recommendation systems, AI affects cognition through a different mechanism: cognitive offloading. When AI can look up any fact instantly, remember everything for you, navigate without your knowing the map, and draft your writing, you increasingly delegate to it the cognitive work you would otherwise do yourself. Some offloading is straightforwardly good. Writing tools that catch errors free mental resources for higher-order thinking. GPS navigation for an unfamiliar city is more reliable than guessing. There is no virtue in doing hard arithmetic by hand when a calculator is available. But there is a version of cognitive offloading that erodes capacity: if you never navigate without GPS, you may lose the ability to build spatial maps of places you visit regularly. If you never write without AI assistance, you may lose fluency in turning your own thoughts into sentences. If you never try to remember before searching, you may lose the practice of retrieval that strengthens memory consolidation. The question is not whether to offload — you will, and often should. The question is which cognitive capacities are worth deliberately maintaining because exercising them matters to the quality of your thinking, your learning, or your life.