But AI is not a tool in this sense, it is an environment which envelops the user and works on us from the inside out while we naively think that we remain unchanged by our use so long as we are using it carefully and intentionally. The care and intentionality is beside the point, and our confidence in such vigilance probably works against us in the long run.
Sacasas’s warning is aimed precisely at people like me: the careful user, the intentional user, the person who thinks he has bounded the machine and knows what he is doing with it. Very well, he has my attention. But a charge this serious requires more than prophetic posture. If my vigilance is part of my delusion, show me the delusion. If my careful use is deforming me, show me the deformation. Tell me what faculty is being weakened, by what habit, through what mechanism, and by what signs I might recognize the damage. Instead, the essay largely lets pugnacity and citation do the work of argument. McLuhan and Illich are summoned like heavy artillery, but the target remains vague. The reader is meant to feel that something deep has been proved because the vocabulary is grave enough. But the key claim, that careful, intentional use is not merely insufficient but self-deceiving, is asserted rather than demonstrated.
The eternal missing question: compared to what? He writes as if the alternative to AI were some richer, slower, more embodied intellectual life waiting patiently nearby. In my case, the alternative is often that the conversation does not happen at all. A stray intuition appears during a hectic week, flickers for a moment, and then disappears under work, fatigue, errands, training, and the ordinary attrition of life. AI does not replace the seminar, the monastery, the long walk with a learned friend, or the silence of the study. Those were not on offer. It supplies kindling for sparks that would otherwise go out.
The question, then, is not whether AI forms us. Of course it does. So do books, notebooks, friendships, instruments, gyms, marriages, and tools of every serious kind. A barbell is not “neutral” in the sense of being consequence-free; used stupidly, it can deform the body. But no sane person concludes from this that barbells are therefore not tools, or that careful use is only fooling oneself. The conclusion is that dangerous tools require form, limits, instruction, humility, and disciplined attention. Sacasas wants to deny the slogan that technology is “just a tool,” but he overcorrects into a different simplification: that because AI is formative, agency is naïve. The most charitable framing might be that the vast majority of people don’t use AI the way I use it. Fair enough if true, but I don’t know how we could quantify a “median use.”