AI-generated “workslop” is destroying productivity?
46 comments
·September 22, 2025ryandrake
noosphr
Just ask an Ai to write how it made you more productive in daily work. It's really good at that. You can pad it out to 1m words by asking it to expand on each section of with subsections.
yifanl
Ways AI have made me more productive: Spellcheck has reduced the number of typos I've made in slack threads between 4 and 10%.
Yoric
« AI has made me productive by writing most of the answer to this question. You may ignore everything after this sentence, it is auto-generated purely from the question, without any intersection with reality. »
thatfrenchguy
It’s kind of a good way to make your business collapse though, because figuring out the kinds of problems where LLMs are useful and where they’ll destroy your productivity is extremely important
cjbgkagh
Just make shit up, or even better have the AI make shit up for you
romaniv
We call these workers “pilots,” as opposed to “passengers.” Pilots use gen AI 75% more often at work than passengers, and 95% more often outside of work.
Identify a real issue with the technology, then shift the blame to a made-up group of people who (supposedly) aren't trying hard enough to embrace the technology. Nice.
paultopia
Is it the “workslop” that is causing the problem, or the slop that companies demand and that passes for work in the first place? Really wanna summon the ghost of David Graeber (“Bullshit Jobs”) here: if you’re a manager who demands your employees to produce PowerPoints about the TPS reports, you probably shouldn’t be surprised when you get meaningless LLM argle-bargle in return.
rickydroll
Workslop production is how we determine who should get a ticket for Ark B.
donatj
My friends job of late has basically become reviewing AI-generated slop his non-technical boss is generating that mostly seems to work and proving why it's not production-ready.
Last week he was telling me about a PR he'd received. It should have been a simple additional CRUD endpoint, but instead it was a 2,000+ loc rats nest adding hooks that manually manipulated their cache system to make it appear to be working without actually working.
He spent most of his day explaining why this shouldn't be merged.
More and more I think Brandolini's law applies directly to AI-generated code
> The amount of [mental] energy needed to refute ~bullshit~ [AI slop] is an order of magnitude bigger than that needed to produce it.
givemeethekeys
The nephew has no programming knowledge.
He wants to build a website that will turn him into a bazillionaire.
He asks AI how to solve problem X.
AI provides direction, but he doesn't quite know how to ask the right questions.
Still, the AI manages to give him a 70% solution.
He will go to his grave before he learns enough programming to do the remaining 30% himself, or, understand the first 70%.
Delegating to AI isn't the same as delegating to a human. If you mistrust the human, you can find another one. If you mistrust the AI, there aren't many others to turn to, and each comes with an uncomfortable learning curve.
zarmin
In the early aughts, I was so adept at navigating my town because I delivered pizza. I could draw a map from memory. My directional skills were A+.
Once GPS became ubiquitous, I started relying on it, and over about a decade, my navigational skills degraded to the point of embarrassment. I've lived in the same major city now for 5 years and I still need a GPS to go everywhere.
This is happening to many people now, where LLMs are replacing our thinking. My dad thinks he is writing his own memoirs. Yeah pop, weird how you and everyone else just started using the "X isn't Y, it's Z" trope liberally in your writing out of nowhere.
It's definitely scary. And it's definitely sinister. I maintain that this is intentional, and the system is working the way they want it to.
matheusmoreira
> He spent most of his day explaining why this shouldn't be merged.
"Explain to me in detail exactly how and why this works, or I'm not merging."
This should suffice as a response to any code the developer did not actively think about before submitting, AI generated or not.
latexr
I think you might’ve missed this part from the post:
> AI-generated slop his non-technical boss is generating
It’s his boss. The type of boss who happily generates AI slop is likely to be the type of person who wants things done their way. The employee doesn’t have the power to block the merge if the boss wants it, thus the conversation on why it shouldn’t be merged needs to be considerably longer (or they need to quit).
bwfan123
> The amount of [mental] energy needed to refute ~bullshit~ [AI slop] is an order of magnitude bigger than that needed to produce it
I see this in code-reviews where AI tools like code-rabbit and greptile are producing workslop in enormous quantities. It is sucking up enormous amount of human energy just reading the nicely formatted bs put out by these tools. All of that for finding an occasional nugget that turns out to be useful.
RobinL
I largely agree. As a counterpoint, today I delivered a significant PR that was accepted easily by the lead dev with the following approach:
1. Create a branch and vibe code a solution until it works (I'm using codex cli)
2. Open new PR and slowly write the real PR myself using the vibe code as a reference, but cross referencing against existing code.
This involved a fair few concepts that were new to me, but had precedent in the existing code. Overall I think my solution was delivered faster and of at least the same quality as if I'd written it all by hand.
I think its disrespectful to PR a solution you don't understand yourself. But this process feels similar to my previous non-AI assisted approach where I would often code spaghetti until the feature worked, and then start again and do it 'properly' once I knew the rough shape of the solution
oblio
https://www.joelonsoftware.com/2000/05/26/reading-code-is-li...
https://www.joelonsoftware.com/2000/04/06/things-you-should-... (read the bold text in the middle of the article)
These articles are 25 years old.
fzeroracer
Sadly, I've seen multiple well-known developers here on HN argue that reading code in fact isn't hard and that it's easy to review AI-generated code. I think fundamentally what AI-generated code is doing is exposing the cracks in many, many engineers across the board that either don't care about code quality or are completely unable to step back and evaluate their own process to see if what they're doing is good or not. If it works it works and there's no need to understand why or how.
jjgreen
Ship it!
Animats
The article as I see it is just one paragraph that end "So much activity, so much enthusiasm, so little return. Why?" Is there more if you're a subscriber to Harvard Business Review?
keyshapegeo99
I had to disable UBO and my VPN's ad blocker. Then the whole piece showed up
kazinator
That light bulb lying in a pool of epoxy resin is cool as hell though. A lot of poetic talent must have gone into the prompt.
varjag
If you were around for the heyday of Markov chain email and Usenet spam this whole thing is familiar. Sure AI slop generation is not directly comparable to Markov process and generated texts are infinitely smoother yet it has similar mental signature. I believe this similarity puts me squarely in the offended 22%.
AlexandrB
It'll be very funny if any AI productivity gains are balanced by productivity loss due to slop - all the while using massive amounts of electricity to achieve nothing.
Bukhmanizer
I don’t think AI slop is inherently mandatory, but I worry that the narrative around AI will devalue engineering work enough that it becomes impossible to avoid.
mallowdram
"transfers the effort from creator to receiver."
AI is functionally equivalent to disinformation as it automates the dark matter of communication/language, transfers the status back to the recipient, it teaches receivers that units contents are no longer valid in general and demands a tapeworm format to replace what is being trained on.
backprop1989
What’s a tapeworm format?
mallowdram
Whatever the training can't assimilate, yet can be transmitted by users as analytic statements.
oblio
I'm not sure I understand this. Maybe an example would help, please?
lazystar
maybe chatgpt can help us understand it better
mallowdram
chat can't grasp what it hasn't been trained to automate.
drivingmenuts
As someone who has spent the last 30 years honing my craft (programming), all I can say is this: ha ha!
BoorishBears
I've spent 16 years and I won't exactly be cheering if we hit a wall with AI.
I love programming, but I also love building things. When I imagine what having an army of mid-level engineers that genuinely only need high level instruction to reliably complete tasks, and don't require raising hundreds of millions while become beholden to some 3rd party, would let me build... I get very excited.
pixl97
Programming is almost never an objective in itself, it's a stepping stone to some other task. It's nice it pays a good living, but I have a feeling that's going away eventually.
dsterry
Implement a no workslop policy. Reputation will take care of the rest. Basically an educational task.
drivingmenuts
And just think of all the money spent on ChatGPT subscriptions. You’re not gonna see that back anytime soon.
My management chain has recently mandated the use of AI during day-to-day work, but also went the extra step to mandate that it make us more productive, too. Come annual review time, we need to write down all the ways AI made our work better. That positive outcome is pre-supposed: there doesn't seem to be any affordance for the case where AI actually makes your work worse or slower. I guess we're supposed to ignore those cases and only mention the times it worked.
It's kind of a mirror image of the global AI marketing hype-factory: Always pump/promote the ways it works well, and ignore/downplay when it works poorly.