[kerblang]

I'm starting to realize that an LLM isn't gonna take my job, but it's beginning to make the job aggravating enough to quit anyhow. So many managers have decided they're going to have an AI Miracle and aren't interested in hearing otherwise, no matter what staff tells them.

[ericmcer]

Unfortunately the big players are pretty entrenched so the degraded quality that appears once AI fails to replace laid off workers will have minimal impact on their bottom line. And the bar for government is literally as low as "Is this such bad UX that it will cause a revolution?".

So why would they care whether its Covid, AI or a Recession that gives them the excuse to do less and less. The system keeps on rolling, the rich get richer, normal peoples lives get incrementally shittier.

[rdtsc]

> So many managers have decided they're going to have an AI Miracle and aren't interested in hearing otherwise, no matter what staff tells them.

Managers' manager convinced them they should expect an AI Miracle. Now your job is to put on a show to pretend to create an AI Miracle so your manager and their manager can pat themselves on the back.

Under enough pressure to use AI people will just produce code as before but LLM-ize it with more comments and verbose crap to look like AI did it. "See boss, I am using AI, so happy you got us this tool".

However, if you do it too well the next step will be "we don't really need so and so, we'll just replace them with an AI agent since it was working out so well".

[AtheistOfFail]

An LLM may take the interesting parts of my job but the parts that suck (dealing with people) will never be taken over by an LLM.

[xenadu02]

Indeed. Uber Eats now makes you talk to an AI bot among other customer-hostile issues. I've largely abandoned them. The last straw was a driver leaving the food at some random house I could not even identify from the picture. It made me wait 5 minutes before I could do anything at all. Then it made me talk to a bot.

When I eventually got it to issue a refund I realized they kept the service fees and driver tip. For an order I didn't even receive!

If that's the best they can do I'll just go pick it up myself.

[JohnTHaller]

Order directly from the restaurant. You'll get better and faster service. And it'll often be cheaper as they have to increase the price on middle men platforms to pay for the fees.

[hsbauauvhabzb]

Phone and chat support has already happened, robotic law enforcement is the future. Now pick up that can citizen.

[Freedom2]

Notably that isn't an accurate reference. The Combine officer is not a robot, so no robotic law enforcement happened.

[hsbauauvhabzb]

There IS robot law enforcement as part of combine, so you’re not even a good neckbeard.

[newswasboring]

The part I don't understand is why can't they wait for the efficiency gains to materialize before firing people? Better pay a few people for a few months extra than be wrong. If AI is going to bring in all this efficiency, this would be peanuts.

[mikkupikku]

Its like the "throw him into the deep end" method of teaching kids to swim. (I don't endorse it, but it has worked for many people.)

[rmah]

Because, for white collar jobs, that is so rare that it's reasonable to say that it never happens.

[monkaiju]

Everyday I am more and more pleased with our company's (or at least our company's tech department) to effectively ban AI.

[orphea]

My manager thinks if we give it a year or two, no one will write code by hand anymore, we will just generate everything from specifications in English.

[CamperBob2]

At the current rate of progress, that is a reasonable thing to expect. But I'd say give it two to three years, myself. This kind of wholesale paradigm shift tends to take longer than you think it will, and then, once it happens, it tends to happen faster than you think it will.

Except for things like hardware drivers, most of the code that will ever need to be written already has been. It will just need to be refactored and recast for new systems and applications, and current-gen LLMs are already extremely good at that.

The line that separates specifications and source code will get increasingly blurry over the next couple of years, eventually reaching a point where it's no longer worth arguing about.

[orphea]

I have plenty of reasons to think it's not going to happen. Not with the LLM-based generation of AI. So let's just agree to disagree :-)

[CamperBob2]

What are some of your reasons? The usual ones don't seem to be holding up well but I'm interested in new insights, certainly. Obviously you disagree with your manager, but that could be due to any number of things.

[watwut]

Does this have anything to do with AI push? It is fairly straightforward that billionaire class cooperating with Trump admin dont want to pay taxes. Republicans want IRS incapable so that tax fraud flourish. Bonus point is that they will be able to pretend worry about it with minorities.

[kerblang]

> Naturally, AI is expected to play a significant role in all this, making people better at their jobs and more end-user-focused, he said.

> However, Pandya said IRS leaders are telling employees that AI won't endanger their jobs.

Not much of trump supporter myself, but I check HN for tech news rather than politics

[tokyobreakfast]

> Republicans want IRS incapable so that tax fraud flourish.

That's wildly hyperbolic.

[selectodude]

They literally put a tax cheat as head of the IRS last year.

People considering speaking frankly about reality as hyperbolic is how we got here.

[astura]

It's not hyperbolic, it's factual.