My personal thoughts on this are
- A good lawyer + AI will likely win in court against a non lawyer with AI who would likely win in court against just an AI
- A good software engineer + AI will ship features faster / safer vs a non engineer with AI, who will beat just AI
- A good doctor + AI will save more lives than a non doctor + AI, who will perform better than just AI
As long as a human has a marginal boost to AI (either by needing to supervise it, regulation, or just AI is simply better with a human agency and intuition) - jobs won't be lost, but the paradox of "productivity increases, yet we end up working harder" will continue.
p.s. there is the classic example I'm sure we all are aware of, autopilot is capable of taking off and landing since the 80s, I personally prefer to keep the pilots there, just in case.
It's also a great example of why tech executives shouldn't be trusted, at all.
"My thing will break our entire economy. I'm still gonna build it, though." - statements dreamed up by the utterly deranged
What if it just makes most jobs worse, or replaces good jobs with more, worse jobs? "Meat robot constantly monitored and directed by AI overlords" is technically a job.
The problem with “AI will replace all jobs” hype, is that it also comes with a flavor of “and we all will do creative work”, while in reality AI replaces all the creative work and people go back to collecting garbage or other physically demanding and mundane jobs.
The analogies to previous technologies always seem misguided to me. Maybe it allows us to make some predictions about the next few years, but not more than that. We do not know when/where we will hit the limits on AI capabilities. I think this is completely unlike any previous technology. AI is intentionally being developed to be able to make decisions in any domain humans work in. This is unlike any previous technology.
The more apt analogy is to other species. When was the last time there was something other than homo sapiens that could carry on an interesting conversation with homo sapiens. 40,000 years? And this new thing has been in development for what? 70 years? The rise in its capabilities has been absolutely meteoric and we don't know where the ceiling is. Analogies to industrial agriculture (a very big deal, historically) and other technologies completely miss the scope of what's happening.
We’ve seen tech completely eliminate jobs like phone switch operators and lamp lighters.
And it’s decimated other professions like manual agriculture, assembly line jobs, etc.
It seems like people are debating whether the impact of AI on computer-based jobs will be elimination or decimation. But for the majority of people, what’s the difference?
The biggest reason there is such a difference of opinion on this is that people have fundamentally different worldviews. If you have bought into the singularity concept and exponential acceleration of computing performance, then you are likely to believe that we are right on track to shortly have smarter-than-human AI. This is also related to just having a technology-positive versus negative worldview. Many people like to blame technology for humanity's failings when in reality it's a neutral lever. But that comes down to the way people look at the world.
People who don't "believe" in the exponential of computing (even though I find the charts pretty convincing) seem to always assume that AI progress will stop near where it is. With that assumption, the skepticism is reasonable. But it's a poorly informed assumption.
https://en.wikipedia.org/wiki/Technological_singularity#Expo...
I think that some of that gets into somewhat religious territory, but the increasing power and efficiency of compute seems fairly objective. And also the intelligence of LLMs seems to track roughly with their size and amount of training. So this does look like it's about scale. And we continue to increase the scale with innovations and new paradigms. There will likely be a new memory-centric computing paradigm (or maybe multiple) within the next five years that increases efficiency by another two orders of magnitude.
Why can I just throw out a prediction about orders of magnitude? Because we have increased the efficiency and performance by orders of magnitude over and over again throughout the entire history of computing.
>> The automation of farm work is the most notable and most labor-impacting example we have from history, rapidly unemploying a huge portion of human beings in the developing economies of the late 19th and 20th centuries. And yet, at the conclusion of this era (~1940s/50s), the conclusion was that “technological unemployment is a myth,” because “technology has created so many new industries” and has expanded the market by “lowering the cost of production to make a price within reach of large masses of purchasers.” In short, technological advances had created more jobs overall
From the late 19th century to the 1940s/50s is 50 years. It's not really reassuring to middle aged workers who lose their jobs to new technology that 50 years later there will overall be more jobs available.
I will likely be leaving tech bc of business execs getting horny and skeeting all over each other at the cost savings they perceive.
The flip side is that now I am using AI for my own entrepreneurial endeavors. Then I get to be the business exec, except my employees will be AI workflows.
And I never have to deal with a business exec every again.
People compare AI to the automation that happened in e.g. car factories. Lots of people were put out of jobs, and that’s just the way things go, they say.
But the difference is that automotive automation does create way more jobs than it destroyed. Programmers, designers, machine maintainers, computer engineers, mechanical engineers, materials scientists, all have a part to play in making those machines. More people are employed by auto manufacturers than ever before, albeit different people.
AI isn’t the same really. It’s not a case of creating more-different jobs. It just substitutes people with a crappier replacement, puts all the power in the hands of the few companies that make it, and the pie is shrinking rather than growing.
We will all pay for the damage done in pure pursuit of profit by shoehorning this tech everywhere.
It's the uncertainty with the transition. Will I, in my mid-career, be able to get one of these new jobs that spawn out as a result of AI, or will I be displaced into something lower-paying as result. TFA kind of just glosses over the people who get displaced in each transition like a footnote.
For SWE, there is a huge misconception that implementation is bottleneck - for mid level to senior it takes much longer to decide what to build than to actually build it. Each additional line of code adds weight to the airframe.
Even if AI can't replace an entire worker, they may still be able to help one worker do the work of two which by itself could lead to massive unemployment (at least in the short term.)
not all of them, but that's not the issue, the issue is 100 engineers + AI replacing 500 engineers
I grew up in the 80s and we were basically taught that at any moment we could see a mushroom cloud and get a minute to say goodbye to everything. As horrible as that was, I think it taught us to deal with uncertainty in a way that maybe other generations didn't get. We learned to accept that we don't always have control over what's going to happen. And in that acceptance, there is a Zen peace.
The truth is, we don't know how AI will evolve. Maybe it will replace all jobs in 10 years. Or maybe never. Or maybe the world will change bit by bit over the next 50 years until it is utterly unrecognizable. Anyone who tells you that they know for sure is selling you something.
If forced to guess, I would say AI is like electricity or the microprocessor: it will change everything, but it will take decades.
Once you accept that things are going to change, it frees you to focus on what's important. Focus on what you can control: your skills, your effort, and your relationships (business and personal).
"Over the weekend I went digging for evidence that AI can, will, or has replaced a large percentage of jobs."
Perhaps the author is curious whether "AI" will replace "SEO jobs", or "web marketing" jobs
If "AI"-generated answers are replacing www search, and even visits to other websites, then perhaps to some extent "AI" will reduce the number of "SEO" or "web marketing" jobs
https://searchengineland.com/seo-opportunity-shrinking-rand-...
The author is a "web marketer" accusing "Tech Execs" of doing web marketing
Marketers trying to discredit other marketers
What could stop AI from doing all jobs?
>the labor displacing effect of technology appears to be more than offset by compensating mechanisms that create or reinstate labor.
I don't buy into this at all:
>Assuming AI will have an effect similar to 20th Century farm equipment’s on agriculture, why will that labor force behave differently to their 20th Century counterparts (and either refuse to or be prevented from finding new jobs)?
Because "farm equipment" can't also perform the jobs it creates. I'm assuming if/once AI can do most current jobs, it can also do most if not all the jobs it creates.
I was wondering if these claims are supported by numbers. Are we seeing a decrease in employment?
I checked the unemployment numbers of US they have a regular trend. But they are very vague and general.
I cannot attribute layoffs at companies like Microsoft to AI because these things happened before many times.
I have a hard time imagining how the current large model could possibly replace all the jobs, given that the models simply retrieve and re-organize human-generated knowledge, and perform limited interpolation when it comes to code generation. We software engineers open source all kinds of cool code to replace ourselves to a certain degree, but without us, who's going to generate new knowledge to feed the models?
they already expecting the same work for less money. and therefore wanting me to work more. i thought it was supposed to make me work less?
I'll create more work no less. I have a client with zero programming experience who is shipping slope UIs and then asking me to fix them. I barely have time as it is for all the low-quality work I have to fix that he sends me daily.
With AI, companies will ship more features, but their competitors will too. This will likely result in a net-zero gain, similar to the current situation.
AI will not replace developers, they will help developers to solve problems faster. But model collapse is real and we have to see how it can overcome this.
"Jobs" aren't people, but these headlines are making that emotional appeal for engagement. AI will transform work like many technologies of the past and most people will be doing different things or the same things differently in a few years. Just like computers or mobile phones.
You just have to look at who benefits from making these claims and keep asking where they're getting their claims/numbers from.
"Oh no, we invented an AI that is so smart we're afraid of it! We need AI safety!" is literally snake-oil sales pitching. Media outlets that give air to these claims are cringe.
AI isn't replacing people. It's being used to replace labour with capital which weakens labour's negotiating power in order to increase the profits of the capital class. Just as other disruptive technologies have been used in the past.
What the Luddite movement shows us is that society needs to prepare for taking care of highly skilled people. What society didn't do back then was take care of people. Textile workers didn't find jobs elsewhere unless you mean work houses. The myth capitalists fabricated around disruptive technologies is that people displaced by these technologies will acquire new skills and find work elsewhere and that these technologies create new opportunities. It doesn't exactly happen that way.
The same could happen here. The wealthy transfer even more wealth from the labour class to themselves and avoid taxation or doing their part to replace the value they've taken from the labour class.
Update: Changed sentence on the "capitalist myth" to explain what the myth is.
Maybe not 100% of every job, but maybe 80% of every job, which means you can consolidate the human-gap-work in 20% of the previous headcount and fire the rest.
When you’re employed, any efficiency gains you get from AI belong to the company, not you.
I find it amusing how you can promise the world then people will just accept a compromise; "well, that's too far, but maybe it will replace juniors", even when that's also an extraordinary claim.
> The AI Fear & Hype Marketing Flywheel
And I suspect the majority of this flywheel is fully manufactured at this point.
Dead Internet Theory is nearly complete.
AI can probably generate music that sounds like Bob Marley but who would pay to watch it perform live? Even if free…
It's not just tech execs doing marketing!
Idiots everywhere are repeating it for them ad nauseum.
Not just marketing, but often trying to cover financial issues by veering into AI hype
I think the AI job extermination narrative has so much traction right now because it's a convenient cover story for layoffs. Every tech CEO is talking about how they're replacing jobs with AI but I don't see stories on HN about how any of these jobs are being replaced with AI.
Thar doesn't mean AI won't replace jobs. I don't know the scale, but it certainly will replace some jobs at least. Just probably not the ones we're currently losing.
The problem with the discourse around AI is that people tend to fall for the extremes.
"AI will replace all programmers within 1 year"
vs
"AI is just another fad like NFTs"
Both sides are very obviously wrong, and the truth lies somewhere in the middle. Most people knowledgeable about AI agree that it will eventually surpass humans in all tasks requiring mental labor (and will thus displace those humans, as using AI will be cheaper than employing people), but no one knows exactly when this will happen. I personally believe it will happen within the next 10 years, but it’s really just a slightly educated guess.
I rarely hear people talk about the fact that demand for new software will increase exponentially as cost to produce software crash. The ratios of devs_per_1k_lines_of_code_in_2022:devs_per_1k_lines_of_code_in_2028 and increase_in_demand:decrease_in_cost are unknown. But what if 1 dev could produce 300% more code, reducing eng cost (assuming same cost per hour per dev) to 1/3 of historical norms. Would that result in an increase of 20% in demand? Probably not, more like 1000%.
If that's the case, there is a large net increase in demand for experienced devs who know how to use AI for coding. Demand will go up massively, I have zero doubt of that, but will AI get so much better that unskilled MBAs are making large complex apps? ¯\_(ツ)_/¯
My take: It won't replace all the SWE jobs. But it _will_ replace many of the entry-level jobs, thereby overtime significantly reducing the number of people in the industry. This is because most companies are focused on short-term cost-cutting rather than training and retaining talent (why spent $$ on that when the talent might then just hop over to another company, especially when what that entry-level person does could instead be done by an AI -- not autonomously, not, but by a more senior dev "overseeing" a number of spawned LLM instances.
https://news.ycombinator.com/item?id=44181342
Hottakes like these are getting retarded.
AI will make it worse - it will replace "easier" jobs, harder jobs will remain, but the bar will be raised and so many people will not be able to have an "office" job.
Related:
The ‘white-collar bloodbath’ is all part of the AI hype machine
No it’s a realistic possibility. Not just marketing.
It may not replace us and it also may. Given the progress of the last decade in AI it’s not far off to say that we will come up with something in the next decade.
I hope nothing will come but it’s unrealistic to say something definitively will not replace us.
Glorified chqtbots. We'll never have AI.
Maybe. But I think this is a useful time to engage in the conversation about what we do as a semi-post-scarcity society and how we should treat people under that framework. If we _do_ replace all the jobs, how should we, as a society, treat each other?
Fundamentally, you either continue down the path of pure capitalism and let people starve in the streets or you adapt socially.
The worse is when tech execs elsewhere actually believe the BS or (more likely) use it as an excuse to put pressure on already overworked employees through layoffs and fear mongering.
This is very real and it's happening now across the industry, with devastating consequences for many (financial, health, etc.)
[dead]
[dead]
[dead]
Of course AI will replace us.
Just like the steam engine did, just like robots did, just like computers did.
Oh, wait.
Not an expert here, just speaking from experience as a working dev. I don’t think AI is going to replace my job as a software developers anytime soon, but it’s definitely changing how we work (and in many cases, already has).
Personally, I use AI a lot. It’s great for boilerplate, getting unstuck, or even offering alternative solutions I wouldn’t have thought of. But where it still struggles sometimes is with the why behind the work. It doesn’t have that human curiosity, asking odd questions, pushing boundaries, or thinking creatively about tradeoffs.
What really makes me pause is when it gives back code that looks right, but I find myself thinking, “Wait… why did it do this?” Especially when security is involved. Even if I prompt with security as the top priority, I still need to carefully review the output.
One recent example that stuck with me: a friend of mine, an office manager with zero coding background, proudly showed off how he used AI to inject some VBA into his Excel report to do advanced filtering. My first reaction was: well, here it is, AI replacing my job. But what hit harder was my second thought: does he even know what he just copied and pasted into that sensitive report?
So yeah, for me AI isn’t a replacement. It’s a power tool, and eventually, maybe a great coding partner. But you still need to know what you’re doing, or at least understand enough to check its work.