Does careful AI use actually protect against reshaping your judgment?

Sacasas opens flatly: "Your AI is not a tool. It is an environment, and you are in it." He is arguing against the common defense that "technology is just a tool — what matters is what we do with it," a framing he calls "the numb stance of the technological idiot" when borrowed from McLuhan. His claim is that AI "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." He backs this with a reported observation from inside a 600,000-person company: "The teams using AI most carefully are the ones losing the ability to tell a good option from a merely safe one. The malformation doesn't skip the diligent. It recruits them" — plus a parallel anecdote from a computer scientist who could "feel the subtle shifts in his own awareness and judgment" while watching the same thing happen to junior colleagues.

The mechanism he gives is McLuhan's medium theory, not a claim about AI's content or accuracy: "the content... is like the juicy piece of meat carried by the burglar to distract the watchdog of the mind," and technology's "effects... do not occur at the level of opinion or concepts, but alter sense ratios or patterns of perception steadily and without any resistance." He extends this with Illich, for whom tools "shape our mind" beneath the level of what we use them for, and whose proposed remedy was not information literacy but an "asceticism... of a sensual praxis" — a disciplined retraining of perception that Illich explicitly linked to the capacity for friendship. Sacasas also cites Charley Johnson's critique of advice to use AI as "a researcher, tutor, and editor at your side": "the medium is doing the work, not the information — which is why you can turn down every suggestion and still end up somewhere you wouldn't have reached alone, feeling the whole time like you walked there on your own two feet." His proposed response is accordingly "not media literacy in a cognitivist mode" but "the training of our perception in an ascetical mode."

This cuts against the vigilance framing in Does AI assistance weaken our brain's ability to think independently?, where a cited survey resolves augmentation-vs-atrophy by saying the outcome "turns on permanent cognitive vigilance." Sacasas's anecdote reports the opposite: the most careful, diligent users were the ones losing judgment fastest, which would make vigilance a thing the environment recruits rather than a lever that controls the outcome. His environment framing also supplies a mechanism for Does AI reshape expert work into knowledge management?: if the shift from creation to curation happens "gradually and almost invisibly," as that note puts it, Sacasas's claim is that this is not a choice experts are making badly but a perceptual reshaping happening beneath the level of choice altogether — which also fits that note's observation that junior workers entering the custodial era never build the reservoir of judgment that older experts have.

The excerpt does not establish any of this empirically — its evidence is one unnamed reader's report from inside a large company, one lunch conversation, and two essays by other writers, not a study of malformation or a test of whether careful use in fact fails to protect users. The McLuhan and Illich citations are analogy and authority, not measurement, and the piece does not specify what "ascetical" perception-training would consist of or whether it would work. The claim is best read as a strong reframing hypothesis — AI as environment, not instrument — that would, if right, explain why vigilance-based remedies in the AI-and-cognition literature keep falling short, rather than as itself a demonstrated mechanism.