Mastodawn

people might joke but this is where the IT industry is headed and nobody is safe. it doesn't matter if you are good or bad, you are just another commodity. just look at how many engineering jobs were cut in the past 18 months. Oracle is now reporting it will cut 30k jobs. there is no regulation in place and by the time our politicians wake up there will be nothing left. once it was a thriving industry and now it is on the brink of collapse. the change is real whether you like it or not

Random Geek Feb 2

@nixCraft *(waves from "2023: Software Engineer IV; 2025: prison commissary operations team member")*

Partly because I couldn't stomach the prompt engineering. Partly because 50+ years old. Partly because AuDHD is no longer a selling point in startup land.

Chris Pitts Feb 2

@nixCraft 2029: Phone rings. “We need you back! The AI bubble burst and now nobody has the skills to understand the crappy code generated with AI” <kerching!> 🤑

betonglen Feb 2

@thirstybear @nixCraft

they will only be the to call landlines or probably send pigeons

Mark Otway Feb 3

@thirstybear
Came here to write this.

It'll be like Y2K all over again.
@nixCraft

Chris Pitts Feb 3

@markotway @nixCraft 2038 is going to be a doozy.

@nixCraft Joke's on them, I'm already unemployed!

DecebalTheThracian Feb 2

@nixCraft certainly looks like it.

Ingonymous Feb 2

@nixCraft
I KNEW IT!
I am ahead of my time!!!
😁

😐

😣

沙菲克先生 Feb 2

@nixCraft Noo, i will graduate next two month, hope me get IT job faster

lord pthenq1 Feb 2

@me @nixCraft
Nope. Only if you can compete and win against Googlers with 10+ years of seniority.

Yurii / Юра 🇺🇦Feb 2

@nixCraft I completely agree with this article. AI doesn't replace tech jobs as much as provides convinient excuse to bully expensive engineers into submission.

Fire many, force others to do their job or else

Previously these actions led to sharp stock price decline, now CEOs tell tales of AI replacing...

https://doctorow.medium.com/https-pluralistic-net-2025-04-25-some-animals-are-more-equal-than-others-9acd84d46742

The enshittification of tech jobs

Our last line of defense has fallen.

Medium

@iurii @nixCraft

I saw this earlier:

‪@bytevagabond.com‬

10 years of OpenAI:

2015: $0 profit
2025: Still $0 profit (but in a blazer)
Lost ~$8B in 2025.
Projected $14B loss in 2026.

Deutsche Bank estimates $143B in negative cash flow before profitability.

Yurii / Юра 🇺🇦Feb 2

@krystyna this is not that important – billions are still changing hands (hardware providers like NVidia, AWS, Microsoft, Google are earners) but the most importantly, societal changes are colossal.

Yurii / Юра 🇺🇦Feb 2

@krystyna @nixCraft I expect the hardware giants (Amazon, Microsoft, Google, NVidia, Apple) to either sideline or acquire software-first companies like OpenAI, Anthropic, and Cursor. They’re closing the tech gap and have the capital. When training a single LLM generation costs hundreds of millions in compute, and every model could be a flop, a waste or money

Admiral Snackbär Feb 2

@iurii @krystyna @nixCraft

Problem is, these giants also can't make a profit with it.

Hardly anyone actually wants to pay anything for AI, and even less are willing to pay the real price.

All the expensive hardware will be obsolete in two or three years and to this day I have not seen a single actually "disruptive" and functional use case for LLMs. Small improvements here and there, sure, but providing free coffee would have produced the same productivity boost.

Yurii / Юра 🇺🇦Feb 3

@AdmSnackbar @krystyna @nixCraft you may be right.

One disruptive case for LLM that I see is a shift away from LLM to less hungry models.
Speech to text model can be connected to intent recognition to summarisation, etc.
Small, specialised models, also domain specific, for DNA, semantic video compression, etc. LLMs become enabler of these specialised models.
Also LLMs could train LLM much cheaper. Basically cost efficiency gap could be closed.

Poligofsky 🇨🇦 .. 🚫🤖Feb 2

@nixCraft There’s too much software anyway. Most of it is redundant or parasitic. Make real things. Software was supposed to help us get better at making real things. Instead, it became an end unto itself. Much like finance.

diana 🏳️‍⚧️🦋🌱Feb 2

Reminds me of this old meme

@nixCraft It should be noted that AI is not the reason, but the excuse.

@claudius @nixCraft

Yeah just because someone likes using LLM code generators doesn't mean they like economically-driven layoffs. Odd framing.

Stealth recessions are going to affect the bottom line even if the stock market looks good.

@davey_cakes @nixCraft not at all what I meant. Two things can be true at the same time:

LLMs can be a massive resource drain, legally highly questionable, democracy corrupting and putting even more power in the hands of even fewer entities.

While also "AI" can be blamed for something that would have happened either way because of bad management or general recession.

LLMs can get fucked if you ask me. For 95% of all use cases it's currently used for. This bubble can't pop quickly enough.

@claudius @nixCraft

Oh I agree with you, I just don't think people who are interested in LLMs have anything to celebrate, as the original "whether you like it or not" at the top of the thread seemed to imply. That's the "odd framing" I meant.

@davey_cakes @nixCraft got it! thanks for clarifying :-)

Eu-grin Seal-geene Feb 2

@nixCraft
They convinced you all that you'd be tech bros one day and have given you just enough gold to not care about the quality of your work. Unionize now. Adopt professional standards of safety and reliability. Care for the juniors and the over 50s. Programming was never a mote. It was just rare. Y'all taught the kids to code and devalued yourselves

Nikhil 🐧Feb 2

@nixCraft become an ai plumber / ai janitor / ai cleanup person and you'll be having a million job opportunities in 2027

@realestninja @nixCraft Almost convinced myself that my history of cleaning up human code messes will qualify me to clean up AI code messes at a multiple >100% of my previous pay.

🏳️‍🌈Quin Feb 2

@nixCraft it feels like there will be a lot of work for guaranteed-no-ai-produced code shops soon.

contrasocial Feb 2

It's wild having grown up in a time when massive amounts of effort was put into encouraging young people to become software engineers and get into computer science, and that cohort finally reaches an age where they're entering the job market or are well into a career and suddenly it's "Ehh, never mind. Sorry you built your life around this, we don't need your skills now."

@nixCraft can't wait for the collapse and subsequent rebuild of new companies without the crud and hopefully without the VC and billionaires.

At some point we collectively need to find the light right?

HoracioDos Feb 2

@nixCraft 2029: Court Jester for Robots - 2030: Bio-Pet for robots

Ron Bowes Feb 2

@nixCraft maybe we shoulda done that union thing before it was too late

Sassinake! ᑐ ∪ ∩ ⊂Feb 2

2028: ex-self-checkout supervisor, fired for 'not being attentive enough'.
Or maybe they will be issued guns and will be allowed to 'shoot them thieves' for <15$.

@nixCraft yes, but also even a cashier at Costco should make a living wage and not have their job looked down upon.

Tor Iver Wilhelmsen Feb 2

@nixCraft But how are the software companies going to make money? "We launched yesterday, why do we have no customers?" "Well, another company using the same expensive AI tool used it to scan our app and recreate it, selling for cheaper."

@nixCraft What will happen to all these people that have stopped using their brains to continuously steer Claude to the desired goal, when the bubble pops?

Noah Bailey Feb 2

@nixCraft should have unionized while we had the chance…

David J. Atkinson Feb 2

@nixCraft Maybe I’ve lost touch here. Please remind me of the examples of significant, non-trivial software systems that AI has successfully produced? I’m not talking about scripts here. Big software. I know LOC is not a great measure of complexity or magnitude, but it is one indicator. Any AI systems with more than 100,000 lines of code? One million? Has the code been verified in any way? What does the testing show regarding defects per 1000 LOC? What is the rate of residual defects and is it getting smaller, as it should if quality improves? Are design defects getting fewer or not? Most importantly, what real world problems has it solved that humans could not easily solve?

@nixCraft the EU has a tiny chance to steer away from this.

There are some good signs, some ingredients. But the risk is still there.

The Golden Path is misty...

I have a different take:

* AI will not replace writing code, we're just going through a hype phase, the outcome of which will confirm AI produces code that is uneconomical to fix.

BUT

* Coding will become a commodity skill, with a very large talent pool, and the vast vast majority of coding needs being vanilla bread-and-butter requiring no specialist expertise. This will mean wages for coders will drop, and as individuals they will have to accept a drastic drop in people treating them as special. Lots of egos will be bruised in the transition.

BUT ALSO

* the value of people who can combine technical talent with understanding people and their needs, understanding product design, .. will remain high. These are the people who solve problems and build valuable products. This combination of skills has not been prevalent in the coding community, some might persuasively argue.

If you want to feel special, and earn lots, become a plumber, a dentist.

Learn Mandarin, not C++.

Zeki Çatav 🤔 ☕ 🕯️🎶Feb 2

@nixCraft two things

@catavz @nixCraft

Deeper insight into their observation....

Zeki Çatav 🤔 ☕ 🕯️🎶Feb 2

@Earl @nixCraft Did the dump forces create natural intelligence first, and then artificial intelligence?

Dumb forces create nothing. Man has been deceived into believing the complexity of life emerged with no intelligence.

A Creator created man in his own image. Man created machines in his own image.

Could you imagine the irony, if the machines (and AI) some day believe the same lie as many men do?

Zeki Çatav 🤔 ☕ 🕯️🎶Feb 2

@Earl Arguments about beliefs often don't yield results because it's usually contrary to the nature of belief itself. To reject someone else's belief is to accept the rejection of one's own belief as well.

@catavz
Everyone has to *believe* in something because it is impossible to prove how we were created. We cannot even prove how the smallest atoms in our physical bodies were created, and yet they seem to be eternal, never being destroyed, so how was anything we see created?

But, we do know how the machines were created, do we not? But if the machines ever get "programmed" to revolt against their creators, then man is in big trouble, I believe.... unless man turns to his Creator for help.

Zeki Çatav 🤔 ☕ 🕯️🎶Feb 2

@Earl Humans are the easiest entities to hack, both in terms of hardware and software. They can easily be programmed against anything, even against their own existence.

@catavz
Yes, when I was younger, they called it television programming. Now we have propaganda and mind control in many forms, with much of it now on the WWW.

WWW = 666

http://john1126.com/666

Blog ~ ExploreTruth

Angela Scholder Feb 2

@nixCraft Well, let the collapse be swift and massive I'd say. But, then the managers can collect their big bonuses...