has work changed?

I think it's fair to say that white collar work is changing. The nature of work, I mean. Obviously it changes all the time — I'm sure those of you old enough remember the introduction of workplace chatrooms — but the AI upheaval has I think been the chang-iest.

There's now a class of software (that your boss is begging you to use!) that can do, in a sense, some parts of your job. They can't do everything, let alone do many things well, but they're good enough. As in: you use them and there are enough people around that someone, somewhere can pick up the slack. Someone in your organization is required to care enough about the services you're providing to make sure the outputs that go into the products are worth paying for. (Presumably.)

AI is another innovation that relieves the laborer of some of their labor. A small part. It also, paradoxically, seems to make said laborer less capable; if you're paying attention to the news, every so often you'll see a study like this one about doctors who perform colonoscopies float across your feeds:

Between Sept 8, 2021, and March 9, 2022, 1443 patients underwent non-AI assisted colonoscopy before (n=795) and after (n=648) the introduction of AI (median age 61 years [IQR 45–70], 847 [58·7%] female, 596 [41·3%] male). The ADR of standard colonoscopy decreased significantly from 28·4% (226 of 795) before to 22·4% (145 of 648) after exposure to AI, corresponding with an absolute difference of –6·0% (95% CI –10·5 to –1·6; p=0·0089). In multivariable logistic regression analysis, exposure to AI (odds ratio 0·69 [95% CI 0·53–0·89]), male versus female patient sex (1·78 [1·38–2·30]), and patient age ≥60 years versus <60 years (3·60 [2·74–4·72]) were the independent factors significantly associated with ADR.

Interpretation
Continuous exposure to AI might reduce the ADR of standard non-AI assisted colonoscopy, suggesting a negative effect on endoscopist behaviour.

The doctors seemed to get worse. This of course isn't the only study, only the first one I found. It appears to make you less effective at the thing you do, the field you specialize in. Which for now isn't a huge issue, if you're not currently looking for work.

There's a lot of talk in elite financial spheres about how workers will be left behind if they don't use the software that makes them worse at their jobs. In fact it was the party line almost since AI's public debut: this technology is the future. Don't get left in the past. And the folks hiring have taken note. Selection pressures have gotten job ads to all evolve the same phrase: "Use generative AI to accelerate [whatever the job in question is]."

It's a funny idea, given that the whole point of these systems is to be consumer-friendly, easy enough that people are betting fortunes a child can get a useful output. Truly a technology out of a cyberpunk future, no? Powerful enough to hack government websites, but controllable by anyone in the world, specialist or not.

That slippage — between how powerful the software at hand is and how simple it is to use — has seemed to put CEOs everywhere in a defensive crouch. They're freaked out. If their rivals use AI to get ahead, what will happen to their companies? And so a hypothetical competitive advantage begets others, even as the purported benefits don't seem to be materializing.

Money is the same everywhere, though: spending more of it in one place means you have less of it to spend in another. More money going to AI means less money available for things like new, unproven hires. You see fewer roles at the entry and mid-levels, and many more in the senior ranks; having a real track record goes further than ever. Meanwhile, if you lose your job, it appears to be harder than ever to get a new one; according to BLS statistics for September '26, 3.133 million people have been unemployed^[1] for 15 weeks or more — out of 7.1 million unemployed workers total. In other words, 44 percent of people currently out of work and looking for it. Anecdotally, just about everyone I know is hanging on to their jobs, even if they're absolutely miserable. They know it's better to have a shit job than to be unemployed in this market.

Anyway, here's one last piece of the puzzle. If a white collar job is hard to find, what do you do? Well, you start a newsletter, if you're a journalist. A PR firm, if you're in public relations. A tech consultancy, if you're a tech worker who isn't really a coder. You become an entrepreneur.

I don't know if that's a good thing, but I do think it's how work has changed in the last few years. I got laid off in August^[2] and it's really struck me how much my slice of the labor market has changed; all the jobs appear to be at firms in various industries — real estate, parking, family offices — that both require the use of AI (because investment!) and are fairly senior. But what's a little more strange is the in-between jobs — those temporary contracts that can keep you afloat month to month as a freelancer — have seemingly disappeared since my last go-round. The work just isn't there, and I strongly suspect there's a two-initial culprit.

The thing that strikes me about this present state of affairs is how it's turned the promise of jobs and business on its head. The reason one took a job, at least before, was security. You got a regular paycheck for regular work. You started a company if you spied an opportunity and had a stomach for risk, the risk being you don't make any money. It's not a good thing to feel forced into.

The social relations that have governed white collar work seem to be changing pretty quickly, from the newly pervasive belief that AI can replace human workers to the idea that training up juniors in your field isn't that important. Of course upheaval is the way of things. Things change slowly, and then all at once.

---

1. Fun fact: the BLS defines unemployed as having looked for a job in the 4 weeks prior to the survey. And so: "Among those not in the labor force who wanted a job, the number of people marginally attached to the labor force decreased by 236,000 to 1.5 million in September. These individuals wanted and were available for work and had looked for a job sometime in the prior 12 months but had not looked for work in the 4 weeks preceding the survey. The number of discouraged workers, a subset of the
   marginally attached who believed that no jobs were available for them, changed little over the month at 414,000." I'm no economist! But those figures would seem to change the picture a little bit. That's another ~ 1.9 million people who want a job but don't have one. ↩︎

2. Okay the story is a lot more complicated than that. Buy me a beer and I'll tell you about it. Also, if you have a job (or writing work)... get in touch. ↩︎