AI is how bosses wage war on “professions”
I'm coming to COLORADO! Catch me in DENVER on Thu (Jan 22) at The Tattered Cover, and in COLORADO SPRINGS this weekend (Jan 23–25) where I'm the Guest of Honor at COSine. Then I'll be in OTTAWA on Jan 28 at Perfect Books and in TORONTO with Tim Wu on Jan 30.
Growing up, I assumed that being a "professional" meant that you were getting paid to do something. That's a perfectly valid definition (I still remember feeling like a "pro" the first time I got paid for my writing), but "professional" has another, far more important definition.
In this other sense of the word, a "professional" is someone bound to a code of conduct that supersedes both the demands of their employer and the demands of the state. Think of a doctor's Hippocratic Oath: having sworn to "first do no harm," a doctor is (literally) duty-bound to refuse orders to harm their patients. If a hospital administrator, a police officer or a judge orders a doctor to harm their patient, they are supposed to refuse. Indeed, depending on how you feel about oaths, they are required to refuse.
There are many "professions" bound to codes of conduct, policed to a greater or lesser extent by "colleges" or other professional associations, many of which have the power to bar a member from the profession for "professional misconduct." Think of lawyers, accountants, medical professionals, librarians, teachers, some engineers, etc.
While all of these fields are very different in terms of the work they do, they share one important trait: they are all fields that AI bros swear will be replaced by chatbots in the near future.
I find this an interesting phenomenon. It's clear to me that chatbots can't do these jobs. Sure, there are instances in which professionals may choose to make use of some AI tools, and I'm happy to stipulate that when a skilled professional chooses to use AI as an adjunct to their work, it might go well. This is in keeping with my theory that to the extent that AI is useful, it's when its user is a centaur (a person assisted by technology), but that employers dream of making AI's users into reverse centaurs (machines who are assisted by people):
https://pluralistic.net/2025/12/05/pop-that-bubble/#u-washington
A psychotherapist who uses AI to transcribe sessions so they can refresh their memory about an exact phrase while they're making notes is a centaur. A psychotherapist who monitors 20 chat sessions with LLM "therapists" in order to intervene if the LLM starts telling patients to kill themselves is a "reverse centaur." This situation makes it impossible for them to truly help "their" patients; they are an "accountability sink," installed to absorb the blame when a patient is harmed by the AI.
Lawyers might use a chatbot to help them format a brief or transcribe a client meeting (centaur)- but when senior partners require their juniors and paralegals to write briefs at inhuman speed (reverse centaur), they are setting themselves up for briefs full of "hallucinated" citations:
https://www.damiencharlotin.com/hallucinations/
I hold a bedrock view that even though an AI can't do your job, an AI salesman can convince your boss to fire you and replace you with an AI that can't do your job:
https://pluralistic.net/2025/03/18/asbestos-in-the-walls/#government-by-spicy-autocomplete
But why are bosses such easy marks for these gabby AI hustlers? Partly, it's because an AI can probably do your boss's job – if 90% of your job is answering email and delegating tasks, and if you are richly rewarded for success but get to blame failure on your underlings, then, yeah, an AI can totally do that job.
But I think there's an important psychological dimension to this: bosses are especially easy to trick with AI when they're being asked to believe that they can use AI to fire workers who are in a position to tell them to fuck off.
