At one side, people are unhappy about AI, at the other side, who of those same people will stop using ChatGPT to write their work e-mails and assignments for them.
It looks like the "car problem" in yet another form. Many people will agree that our cities have become too car-centric and that cars take way too much public space, but few will give up their own personal car.
> who of those same people will stop using ChatGPT to write their work e-mails and assignments for them
Me. I never use AI to write content that I put my name to. I use AI in the same way that I use a search engine. In fact, that is pretty much what AI is -- a search engine on steroids.
Good. I can believe that a few people are principled enough, but principled people tend to be in a minority, regardless of the topic.
I am also a bit afraid of a future where the workload will be adjusted to heavy AI use, to the degree that a human working with his own head won't be able to satisfy the demands.
This happened around the 'car problem' too: how many jobs are in a walkable / bikeable distance now vs. 1925?
I don't think AI is comparable to cars. The problem with cars is that they necessarily use the commons. The more roads you build, the less space you have for trains, parks, housing, etc. AI isn't like that. I can continue to think for myself and look for ways to add value as a human even if everyone around me is using AI. And if that fails, if I can't find a way to compete with AI, if AI really is capable of doing everything that I can do as well as I can do it, why would I not want to use it?
Tell that to anyone who was hoping to upgrade their RAM or build a new system in the near future.
Tell that to anyone who's seen a noticeable spike in electricity prices.
Tell that to anyone who's seen their company employ layoffs and/or hiring freezes because management is convinced AI can replace a significant portion of their staff.
AI, like any new technology, is going to cost resources and growing pains during its adoption. The important question which we'll only really know years or decades from now is whether it is a net positive.
Nope, we have a lot more sprawl. Look at the old maps of cities and compare them to the current ones.
In Ostrava, where I live, worker's colonies were located right next to the factories or mines, within walking distance, precisely to facilitate easy access. It came with a lot of other problems (pollution), but "commute" wasn't really a thing. Even streetcars were fairly expensive, and most people would think twice before paying the fare twice a day.
Nowadays, there are still industrial zones around, but they tend to be located 5-10 km from the residential areas, far too far to walk.
Even leaving industry aside, how many kids you know walk to school, because it is in a walking distance from them?
I never use AI to write an email, and if I ever found out a coworker was using AI to sent emails to me I would never read those emails. It would be a tacit admission that the coworker in question did not have anything worth actually reading.
I am a fairly prolific writer, having published ten books since 2018 and averaging some three articles per week, all of that next to my programming work.
But I understood quite early that I am a fluke of nature and many other people, including smart ones, really struggle when putting their words on paper or into Word/LibreWriter. A cardiologist who saved my wife's life is one of them. He is a great surgeon, but calling his writing mediocre would be charitable.
Such people will resort to AI if only to save time and energy.
And the LLM can parse out total garbage in and understand the intent of the writer? I know when I'm vague with an LLM I get junk or inappropriate output.
As an optimist I would say that it could be better at teasing out your intent from you in an interactive way, then producing something along those lines. People aren't ashamed to answer questions from AI.
That might drift in the future. I've actually found myself leaving small errors in sometimes since it suggests that I actually wrote it. I don't use literal em-dashes -- but I often use the manual version and have been doing so much longer than mainstream LLMs have been around. I also use a lot of bulleted lists -- both of which imply LLM usage. I take my writing seriously, even when it's just an internet comment. The idea that people might think I wrote with an LLM would be insulting.
But further and to the point, spelling / grammar errors might be a boutique sign of authenticity, much like fake "hand-made" goods with intentional errors or aging added in the factory.
Unless you are using a proprietary, dedicated grammar checker, auto grammar check is far from perfect and will miss some subject-verb agreement errors, incorrect use of idioms, or choppy flow. Particularly in professional environments where you are being evaluated, this can tank an otherwise solid piece of written work. Even online in HN comments, people will poke fun at grammar, and (while I don't have objective evidence for this) I have noticed that posts with poor grammar or misspellings tend to have less engagement overall. In a perfect world, this wouldn't matter, but it's a huge driving factor for why people use LLMs to touch up their writing.
I started at a new job a few months back and I got an obviously AI-written reply to my manager's "welcome" email from some contractor type person who got CC'd on it. Fortunately I don't have to interact with the bozo in question, but it was really offputting.
I actually think that AI is a great use case for writing emails, starting from a draft or list of what you want to say and getting it polished to a professional tone. You need to prompt it correctly, review and iterate so it doesn’t become slop, but very useful.
OTOH, I’d never use it to write emails to friends and family, but then I don’t need to sound professional.
I think you have a good point, but are getting a lot of pushback because of your example. Most AI-hostile people won't use ChatGPT directly but are still happy to use a lot of modern AI features/products such as speech-to-text, recommendation engines, translation services, et cetera.
OK, what sort of systemic change you propose? Note that bans on anything digital are really hard to enforce without giving law enforcement draconian powers.
Yes, that's the meme. "We should improve society somewhat" doesn't mean the peasant has actionable proposals, only pointing out there's a problem.
My comment was instead highlighting how your analogy to the "car problem" might be right, in that where we see big shifts to reduce the car problem, like in Paris, it comes from systemic changes from a car-centric form to a diverse transit form, rather than an individual choice model.
My go-to these days is to heavily tax the rich, place a staggering tax on the the superrich, introduce meaningful UBI, put strict controls on housing rentals, etc.
Why waste time using ChatGPT to write work email slop when you don't need to work?
I presume the student is using ChatGPT for assignments in order to get the credentials (a degree) needed for a job - while companies off-load their training costs onto young people, who are then encourage to go into debt, resulting in a mild form of debt bondage.
Reduce the need for a job, so the students who go to college are more likely to be those who want the personal education, rather than credentialism.
But hey, I'm just a peasant programmer saying there are flaws, and we should do something about it. Talk to an actual expert, not me.
Those experts (I hear them on podcasts) will also say things like having strong consumer protection laws so people aren't forced to deal with AI (and human!) sludge.
These new urban systems are simply a way to cram as many people into a small boxes as possible and make citizens culturally flex with their bicycle life and not just seem like a poor peasant. Few give up their personal car because of decades of entrainment. I just think for better or worse, North America is always going to come out with the most selfish (for better or worse) system.
It can be clean tech but we need it to be personal or else we feel like we are declining in standard of living. They don't struggle with these issues in Europe or Asia because Europe and Asia are fundamentally different societies. I don't really see any other way around this dilemma.
blaming the individual instead of the system is a sign of shillbottery
i'd give up my car tomorrow if we had better rapid transit in these parts. And they're working on it, but it takes billions and decades. And I need to drop my kids off at school tomorrow...
It's much worse than "designed for cars." It's more like "not survivable without a car." It's the same with apps on my phone. I don't want to use them, but sometimes there simply is no alternative in today's world.
We may end up building a world where AI is similarly necessary. The AI companies would certainly like that. But at the moment we still have a choice. The more people exercise their agency now the more likely we are to retain that agency in the future.
I lived in Prague, whose center is medieval and the neighbourhoods around it pre-1900, and even though what you say is true (fewer people drove everywhere), the streets were still saturated to their capacity.
It seemed to me that regardless of the city, many people will drive until the point where traffic jams and parking become a nightmare, and only then consider the alternatives. This point of pain is much lower in old European cities that weren't built as car-centric and much higher in the US, but the pattern seems to repeat itself.
Helsinki made a major push to reduce cars to get to Vision Zero and succeeded in no car fatalities in 2024. It’s now hard to get a taxi and you’re expected to walk / other transport it’s a little bit annoying but worth it
The comment explicitly mentioned "cities". Of course rural and suburban areas don't make it practical to be without a car, but many people in cities could use public transportation but handwave it as beneath them or dangerous or unreliable. When in reality it works just fine. Car travel has its own tradeoffs that can be just as easily exaggerated.
> At one side, people are unhappy about AI, at the other side, who of those same people will stop using ChatGPT to write their work e-mails and assignments for them.
As Newsweek points out*, the people most unhappy about AI are the ones who CAN'T use ChatGPT to write their work e-mails and assignments because they NO LONGER have access to those jobs. There are many of us who believe that the backlash against AI would never have gotten so strong if it hadn't come at the expense of the creators, the engineers, and the unskilled laborers first.
AI agents are the new scabs, and the people haven't been fooled into believing that AI will be an improvement in their lives.
It looks like the "car problem" in yet another form. Many people will agree that our cities have become too car-centric and that cars take way too much public space, but few will give up their own personal car.