Live data from Hacker News

AI will not replace software engineers (hopefully)

medium.com

71–76 of 76 posts

Re: AI will not replace software engineers (hopefully)

#71

The tractor analogy keeps coming up in these threads, and I think it's actually more pessimistic than people realize. Tractors didn't just change farming. They emptied entire regions. What saved the people (not the communities) was that other industries absorbed them. Factory work, services, construction. The question for software isn't whether AI creates efficiency. It's whether there's somewhere else for displaced…

The tractor analogy continues to work. Take a look at right to repair.

I think the main "concern" is that Senior devs, code, essentially the entire current working body of programmers is going to bootstrap AI, and once they're gone, they'll be no one to replace them. And at that point, there's no fall back system.

Re: AI will not replace software engineers (hopefully)

#72

The tractor analogy keeps coming up in these threads, and I think it's actually more pessimistic than people realize. Tractors didn't just change farming. They emptied entire regions. What saved the people (not the communities) was that other industries absorbed them. Factory work, services, construction. The question for software isn't whether AI creates efficiency. It's whether there's somewhere else for displaced…

I thought like you 5 months ago. I've come to the conclusion that we won't be replaced, because a majority of our work is to split up business questions, probe, ask around for other people's knowledge, assemble, build a plan etc. AI only knows what it knows, it doesn't go after the unknowns. It is reactionary in its nature. Now, let's say something happens and I'm wrong: let's think about that. The AI can do stuff li…

This is also been my cope.

By the time software engineers are fully automated, then executive assistants, accountants, business development people, marketers, administrative staff, researchers, and HR will also have been fully automated. At that point we have a revolution on our hands and not having a coding job will be the least of your worries.

(or it happens slowly enough that we have time to adjust)

Re: AI will not replace software engineers (hopefully)

#73

Earlier quoted context omitted.

You can't make money playing a guitar and warbling into a microphone anymore either. Hasn't stopped my local bar from hosting local musicians who are in it for the craft.

I don't think that's really an apt comparison. Most software people who are in it purely for the craft are deep in open source, and open source doesn't necessarily build the most user-friendly tooling. If it did, you'd see apps like GIMP seriously challenging Adobe, etc. So much of the software layer built for enterprise and consumers relied on VC. I can only speak for myself, but I'd be willing to bet that 95+% of d…

I'm willing to be that 95% of developers you've worked with are, but how representative that is of the broader group is indeterminate. Apple is full of developers who give a shit about design, or at least, had to pass an interview that pretends that they pretend to give a predend shit... look, at some level you just have to buy that people are who they say they are. Even if you're in it for the money, you're going to conferences and speaking about it to people, so it's you.

Re: AI will not replace software engineers (hopefully)

#75

The tractor analogy keeps coming up in these threads, and I think it's actually more pessimistic than people realize. Tractors didn't just change farming. They emptied entire regions. What saved the people (not the communities) was that other industries absorbed them. Factory work, services, construction. The question for software isn't whether AI creates efficiency. It's whether there's somewhere else for displaced…

What I do not understand is why is it that software engineers are so afraid? I have heard from so many other white collar people that AI already changed their job entirely (technical salesmen, translators, designers, government researchers, the list goes on), yet it is the software engineers that I hear the most noise from. Software engineering is one of the most intellectually demanding categories of white collar wo…

it's because software fundamentally has specific, measurable outcomes. it has to, that's why we build things: to do things. and in computing, necessarily, the things you do can be computed, which means they can be checked for correctness and reproducibility

much of an SWE's role is to artfully construct complex abstractions of reality in such a way that it can be understood by others, with the intent that it can be fixed or improved on in the future.

but if you don't _need_ artful construction in the first place, because you can just brute force AI slopped assembly language that's impossible (for MeatI) to comprehend, and the same AI is sufficient at improving and fixing it, then why do you need an SWE to bring artfulness to the table for you?

we're a long way off from this, ultimately, but imho, this is the broad strokes of it

Re: AI will not replace software engineers (hopefully)

#76
post #75

Earlier quoted context omitted.

What I do not understand is why is it that software engineers are so afraid? I have heard from so many other white collar people that AI already changed their job entirely (technical salesmen, translators, designers, government researchers, the list goes on), yet it is the software engineers that I hear the most noise from. Software engineering is one of the most intellectually demanding categories of white collar wo…

it's because software fundamentally has specific, measurable outcomes. it has to, that's why we build things: to do things. and in computing, necessarily, the things you do can be computed, which means they can be checked for correctness and reproducibility much of an SWE's role is to artfully construct complex abstractions of reality in such a way that it can be understood by others, with the intent that it can be f…

How is software verifiable whatsoever for correctness? Ironically hardware is

I mean pure functions maybe verifiable for correctness, but I don't know how you verify facebooks codebase is 'correct'

Maybe try to run a maximizing function on the stock value, very unfeisable ofcourse

Post reply on HN