If you were an average engineer destined to work in cookie-cutter roles in tech consultancies, would you still pursue a tech career, knowing that your job will sooner or later be taken over by AI?
Remember that careers last 30 years, and if you're the average engineer, you're more likely to fall into the 80% of the pareto half than the 20%. Can you be confident that AI won't displace you in 10, 15, or even 20 years?
This doesn't apply to Stanford grads destined for high-tech roles in FAANG tier companies. I'm taling about the Idaho State grads who will work for IBM and Accenture and likely won't make the $500k salaries to retire from their tech roles in 10 years.
As long as we have human using computers, we'll have more skill humans helping the humans.
We were told that inventing web search engines and giving kids ipads would completely eliminate IT. We see how that turned out LOL.
From a financial standpoint, as long as the top 1% of "IT skills" population can make the other 99% of the company more than 1% more profitable, they'll never get fired, well, completely anyway. At big companies its a pretty low bar to surpass, and small companies can contract with MSPs.
From a cultural / psychology standpoint, its pretty deeply ingrained that some folks are dumb/lazy/ignorant and thats OK, some folks have intense learned helplessness as a character trait, and some folks get really excited about forcing others to "do their work for them" WRT primate dominance. We as a culture will never avoid the appeal of BYOD, even with a miracle of 100% BYOD in IT, we as a culture would STILL hire cloud admins just so the job can be delegated if nothing else.
The pay in IT in most of the country is ... microscopic compared to software dev. New grads or people with nothing but an A+ are not getting new Stanford grad software dev salaries, its more like $22 to $25/hr in most of the country for a new help desk jockey. And it's pretty boring work for 99% of the employees 99% of the time. However, if chatgpt can fizzbuzz better than humans, doing "IT stuff" beats starvation for us computer people.
Its also worth pointing out that AI as a tool is just biased single answer web search. Web search more or less resulted in more employment. Seems likely this variation on the theme will result in more employment.
Dentists are already 3:1 dental hygienists these days and I would expect AI can make that ratio 5:1 or 6:1. Technicians will also be enabled to be more effective with AI (faster diagnoses, remote diagnosis, predictive ordering so you don't have to come back twice with a special part, etc...). Also AI will drive higher reliability of the underlying equipment so the number of technicians needed will be lower.
If AI becomes capable of replacing the bulk of software developers, many trades will be replaced by a guy earning a little over minimum wage wearing AR goggles with an AI showing+telling them what to do and correcting their mistakes.
No need for the AR goggles, just hire day labor to work for a skilled tradesman as a labor multiplier, or try the old fashioned "here's a mobile phone do what your boss says"
In the trades themselves, we've been doing this for centuries... have a master mason supervise the journeyman masons supervising the manual day labor. AI will not be able to compete with experienced personnel. In that field they've already dumbed down the entry level requirement to "pick up a brick and move it there" and it can't go much lower.
On the other hand, if I were manufacturing survey equipment I'd worry about AR goggles as the competition.
Medicine, law, accounting, professional engineering. Notice something in common with all of these?
With the exception of law, automation has already arrived in all of these fields (to accounting first, 1-2 generations ago) and there are now more people in each field rather than less.
Answering my own question: each has a comfortable regulatory moat and competent professional society to defend it.
Yeah, but only certified humans can legally practice law. And they're also the group of people with the best ability to prevent changes in that legislation
I'm not sure on that. the the pieces seem to be there as far as hands that can manipulate rubix cubes and stuff, catch baseballs, boston dynamics robots agility, other humanoid robots.
Do I think white collar jobs go first, yes. also, the dynamics who buys all the stuff that blue collars workers install and build? without a white collar workforce the demand for that goes way down, and the supply of workers goes up.
Not saying it's going to happen just think other jobs aren't that far behind.
Perhaps in the long run everything will be automated, but in the meantime there will be a progression of what AI can do, and software engineering might be "lower-hanging fruit"
Any career with strong regulation is going to take longer to automate. Hands-on work as well.
For example, there is strong regulation preventing computers from practicing driving/trucking/delivery, law, medicine, nursing, physiotherapy, and certified engineering
Also, build me a robot that can do the following. I'll wait: construction, woodworking, metalworking, glassblowing, painting (painting houses), electricking, plumbing, delivery, service industry, cooking.
If you’re entering college right now, you’ll be around 33 in 15 years.
We have no way of knowing what this tech will be like in 15 years.
The iPhone was released about 15 years ago. Even the most prescient of us couldn’t have predicted how fundamentally that would change life, culture, work and business.
Lol, I talked about very short snippets and you counter my argument with a video showing very short snippets.
If I wasn't scared before, I'm even less scared now.
But if all you do is write trivial code anyone can learn to write in 1 week, with extremely detailed instructions, perhaps your job is in danger indeed. That's not what a software engineer does anyway.
According to HN, technology never improves. The first version of something is always the way it will ever be. OpenAI hasn't improved at all from GPT-3 to GPT-4, and all the billions flowing into this field will cause zero improvements in the next 30 years.
This place needs to change its name to LudditeNews.
The main concern of the luddites was not "technology = bad" like most popular caricatures of them, but that their livelihood was obliterated by the first wave of factory owners taking all the profits that were previously shared with the workers for themselves. Really they were pushing in the same direction as those who today are calling for machine taxes or UBI.
Ehhh I'd argue they had more in common with socialists/communists, they were literally destroying the means of production in worker solidarity to improve their condition and bargaining ability. Marx even refers directly to the situation of the guilds in industrialization.
The belief is that new means of production should be embraced, owned, and operated by workers to benefit the many and not just the few exploitative capitalists (which are those who own means of production).
Instead of exploiting lay-offs and long shifts, and disregarding safety to save the capitalist money, the factory would instead be built and operated for the benefit of workers and society, prioritizing safety and sharing shorter shifts among a larger workforce to preserve jobs.
From a socialist/communist perspective machine taxes and UBI are like cheap bribes to avoid the issue and preserve the exploitative system in place. The luddites probably would have felt the same had they been offered a pity sum or machine tax from factory owners to "offset" their condition.
boggles my mind that we had a tool that was smart when it was first launched in November 2022, experienced a massive generational improvement in the next iteration (GPT-4) within 4 months...and people on HN are ready to assert that this is the best it will ever be.
It improved drastically in quality within 4 months, yet for some reason, the nex 10 years will see 0 progress in a field with literally tens of billions and intellectual capital flowing in.
Most revolutionary technical changes like this tend to hit a plateau in progress right after they hit the top of the hype cycle. It’s unlikely that AI/ML will be an exception.
Not bad news for anyone involved in this work, though. That period is when the real business and societal changes happen. It’s also when most of the money gets made.
People and businesses need time to test new technology and adapt it so that it works reliably for their specific use cases. GPT and other LLMs are no exception.
Again, the point is that careers are built over decades, and this proof of concept tech is already good enough to do the job of a coding intern.
That might not be a threat to you if you’re a 35 year old developer, but if you’re an 18 year old exiting high school, there’s a non zero chance that by the time you’re 35 - 17 years - the tech will be good enough to replace you.
The hype ratio between GPT2 and GPT3 is much higher than the hype ratio between GPT3 and GPT4. If we project this into the future then the hype ratio will be even lower between GPT4 and GPT5. The key innovation was RLHF and everything since then has been incremental improvements.
>knowing that your job will sooner or later be taken over by AI?
There's a few layers that make me not worried:
- automation will shift jobs
- an engineer is a generally smart employee (yes, even the average engineers), so the company will be a sinking ship if they have to eliminate jobs and get rid of people instead of moving smart people around
- even if you do end up jobless, as a generally smart person, you can learn a new skill. Plus, now AI can help you learn more quickly.
The catch is, of course, predicting which next job won't be replaced by AI.
Remember that careers last 30 years, and if you're the average engineer, you're more likely to fall into the 80% of the pareto half than the 20%. Can you be confident that AI won't displace you in 10, 15, or even 20 years?
This doesn't apply to Stanford grads destined for high-tech roles in FAANG tier companies. I'm taling about the Idaho State grads who will work for IBM and Accenture and likely won't make the $500k salaries to retire from their tech roles in 10 years.