This is what I think is the solution - I've tried it a few times with good results.
1. We need our own project or task plan. A simple forward-thinking TODO list is good enough. This needs to be prepared before the day begins. Daily.
2. Focus and limit scope. Focusing on doing ONLY what's on the list is the discipline I (we) lack and which we need badly.
3. Side quests are killing us. Allowing ourselves to veer off the planned daily tasks should require a complete away-from-the-keyboard planning session with pen and paper. Print it out but plan it away from the LLM. Use the LLM to summarise the problem, by all means, but do the thinking away from the LLM. The current TODO list needs to be the main input of such a deliberation. The output is the new TODO list. The objective of this away-from-keyboard exercise is simply to maintain control of the next steps.
On days where I worked with my TODO lists, I clocked out and joined the family for activities, slept sooner and better. On days I lost control of that list, the exhaustion mounted almost on the same day.
Make a finite list for the day, tick it off and go spend time with the missus, the kids or the dog. 2c.
Agreed. We're still bottlenecked when it comes to feature work because we need to think of the ideas, user test, talk with stakeholders etc. But somehow inventing loads of tasks to feel busy anyways. Those tasks are the side quests, things we normally wouldn't take on. It feels great to finally squash that bug or do that rewrite. But you're adding stuff onto everyone else's plate as well. Discussions, reviews etc, a constant barrage of new things to consider.
I like this and then I also timeblock so I write out manually on paper every 30 minute increment, and then as the day goes you mark what you did in the previous 30 minute block. Doesn't need to be perfect but it kind of anchors you when at a desk and then it's interesting to look back at the end of the day and see how productive or unproductive you were.
I think AI has also devalued a lot of work that previously might have been "mundane" but still felt like an accomplishment once you got it done. But now you can do that work to an ±80% quality level of the old, manual way, but in 1% of the time.
So the result is noticeably worse, but you basically must do it, because the efficiency gains are inescapable. This makes the work seem even more pointless, adding to the feeling that you're working more and accomplishing less.
Aluminum used to be more expensive than platinum. Now we crumple it up and throw it in the bin. AI has done the same thing to the cost of a line of code.
> Aluminum used to be more expensive than platinum. Now we crumple it up and throw it in the bin. AI has done the same thing to the cost of a line of code.
And hopefully it will to that to all labor, so we can finally crumple up the workers and throw them into the bin too!
And those greedy workers shouldn't complain, they should be happy about the change! Now they can have as much of the stuff we throw in the bin as they want!
And a corollary being that without any time spent on the "mundane", almost every single aspect of the work becomes the genuinely consequential decision-making steps.
I see this directly. I'm an engineer but part of that is doing non-engineer work like making powerpoints, whitepapers, and generic work that's necessary but that I would delegate to someone else if I were high enough on the totem pole.
AI gives me the ability to do that stuff faster so I can focus more on the 20% and the quality of what I'm doing, not the nuance of formatting a document.
AI makes up for a lot of crap tooling, but is often worse than good tooling. In this case "good tooling" would be a document template or formatting language that is easy/fast to use. What I have seen AI do so far if you try to one shot a PowerPoint deck is underwhelming to say the least. (To say more: it looks like a word cannon was used to splatter text on slides with random positions, colors, and fonts.)
Just skip the powerpoint if you can't be bothered to create it. It's painfully obvious for anyone listening that the presenter didn't come up with or think through what they're talking about.
Not always an option if you're being required to make a PowerPoint. The content doesn't need to be AI generated. Just the organization of slides is a PITA and not a good use of my time.
Khm, ackshually... any documentation work is best not left to LLMs. They are excellent at generating many words with little value. Even sketching a doc and running through an LLM although a bit better than just a prompt "generate docs from this code" still suffers from similar problems. I prefer to write briefly, but doing it myself.
OTOH - formatting and other stuff not related to content, this is great to delegate, even though the results are often mediocre, but I really don't care about that at this point.
The opening premise baffles me. What kind of illusion must one be under to believe that employers would ever pay you the same amount of money to do less work?
Realistically there were only ever two possible outcomes: you do more work for the same amount of money, or you’re one of the ones who were laid off and your work was transferred to the former.
The only way it will result in less work and the same or better quality of life is if you’re self-employed or are employed by a compassionate human. Sadly that does not apply to the majority of people.
> The opening premise baffles me. What kind of illusion must one be under to believe that employers would ever pay you the same amount of money to do less work?
It seems like a pretty common misunderstanding in rank-and-file AI-boosterism. I'm guessing that reflects their initial experience, before their employer adjusts their expectations, extrapolated out to infinity. Your correct understanding makes AI sound like a bad deal for workers (and it is), so my theory AI-enthusiastic boosters have to engage in a degree of reality distortion to keep up their enthusiasm.
On the plus side for us worker bees, this is why I never really bought in to the narrative that our jobs were inevitably doomed. A question I've posed a few times: Consider two reasonably similar competitors with 100 engineers two years ago. Suppose AI really does accelerate development 10x times. It doesn't (at least not yet) but the larger number helps bring the question into focus. One company fires 90 engineers and does the same work with the remaining 10. The other company keeps all its engineers and does 10 times the work.
Who wins in the market?
It's not a uniform answer. There are times and places where the first one wins, as well as a mixed strategy. But I think broadly speaking the majority answer is the second.
>On the plus side for us worker bees, this is why I never really bought in to the narrative that our jobs were inevitably doomed. A question I've posed a few times: Consider two reasonably similar competitors with 100 engineers two years ago. Suppose AI really does accelerate development 10x times. It doesn't (at least not yet) but the larger number helps bring the question into focus. One company fires 90 engineers and does the same work with the remaining 10. The other company keeps all its engineers and does 10 times the work.
Who wins in the market?
The one who understands that "10 times the work" is not necessarily absorbable by the market, and adjust to whatever lesser value still works, and still fires a lot of the 100.
> A lot of the fired then make a Discord and casually replicate the company's product.
And what's going to be the result of that? Can the market really support that many competitors comfortably, and/or are you setting yourself up for business failure or Chinese-style involution and 996?
See, I want a comfortable happy life, not a stress-filled one of maximum market efficiency. AI + capitalism doesn't look like it's going to provide the former.
The market can't support it. The value of software will trend towards zero.
The less society spends on developers, the less valuable software is. If the day comes that we employ zero developers, then software will be worth zero dollars.
And yet, software will still be fulfilling an essential part of society, it just wont be a thing people are willing to pay for. So AI will still be around, and will still be essential (because somehow we have to maintain all this software), and AI will get better and better.
Eventually any knowledge work will be worth zero. Including the C-suite--especially the c-suite!
No, what they will do is redefine (again) what is worth paying money for.
We don't need to buy movies, eat avocado, drink wine, wear new clothes, drive shiny cars, etc. We do it because we have been conditioned, over the centuries, to desire what rich people would want. Every time something becomes cheap (e.g. having a watch), capital owners simply redefine it so that it can be sold again as expensive (e.g. having a luxury watch).
> Eventually any knowledge work will be worth zero. Including the C-suite--especially the c-suite!
1. I don't think the "c-suite" is knowledge work.
2. If "any knowledge work will be worth zero" what will workers do? Manual labor? That's a horrible regression.
3. I think the direction AI is taking us is world where the only thing that will be worth anything is ownership (basically being rich), and you either own stuff or you don't. And all the ladders to get from no ownership to ownership will be pulled up.
There’s a flaw in your logic. The value of software doesn’t derive from how much you spend on developers. It’s a function of what customers are willing to pay for the service/product.
If there is no market fit, 100 man-years is worth $0.
On the other hand, if 10K customers are willing to pay $15/month for a tool that took 30 man-months to build you have a handy $150K/month revenue stream.
Most people aren’t interested in vibing and maintaining their own bespoke apps.
But also they probably want to use this app on their phone and other devices. So the either have to make it a web app, become a server admin and pay for hosting. Or they have to buy a Mac and pay for the developer program and hope they pass App Store review so they don’t have to reinstall every days.
I could imagine that someone's plan could be like this: "Our current company has more managers than developers, and the software development is very ineffective because we spend a lot of time in meetings, doing various bureaucracy, and changing things over and over again, because the managers insist on making technical decisions, etc. We could do the same thing with a fraction of the workforce, and also work from home instead of renting an expensive place in the middle of an expensive city. Then our product could be cheaper and we would could still work less and have higher salaries."
Whether this is realistic or not, I can imagine some people actually believing it.
This is why recessions often lead to a burst in new companies being founded.
One of the experiences that stuck with my in my early career was when my coworkers heard talk of layoffs at their startup that was acquired by bigcorp, one of the first reactions was “maybe we should just do another startup.”
I was taken aback at how casually that idea was thrown around. They didn’t even have a specific idea, just the casual thought that they could just start another company.
>I was taken aback at how casually that idea was thrown around. They didn’t even have a specific idea, just the casual thought that they could just start another company.
That's how a lot of companies are made, even huge ones. Not everything comes from a specific focus, it's just "let's see what works to build a business that can make us money".
Some companies start with a temporary idea and later pivot to a new one.
For example, I know a few companies which at the beginning were just body shops, but the long-term plan was to develop their own projects after they figure out what exactly their customers need most.
It's the VC/Silicon Valley mindset. They want to be able to call themselves "entrepreneurs" and "founders"; the product has little importance compared to that. It goes some way to explaining why the vast majority fail.
>A lot of the fired then make a Discord and casually replicate the company's product.
And they find out replicating the product means little, it's the customer base, industry reputations, connections, support, ecosystem, and stuff, that matter.
And then the business learns how many of their customers were only there because of the sales rep, and jump ship to go to whatever vibe coded app the sales rep makes to specifically handle their "unique" environment.
Companies that conduct these layoffs don't just lay off developers. The tend to also lay off support, sales, legal, etc... Especially support, a chatbot is all you need right?
And that's how that "layoff Discord" can produce something that can matter to customers. It will be a smaller company, and small companies have a serious advantage when it comes to support: you can get an actual human, maybe even the main dev, he can actually do things rather than transferring you from chatbot to bureaucracy. It doesn't scale, but it may be enough to get a competitor started.
As for connections, you may have inadvertently let go of the guy who has the connections. And you may have fired a perfectly functional team, ready for from clever entrepreneur with connections (who may be one of the fired) to pick up. It is so much of a recognized problem that some big companies hire people just so that their competitors can't have them.
> some big companies hire people just so that their competitors can't have them.
True. Also true that this was ZIRP phenomenon predominantly in big tech. And once mass layoffs started in big tech most of those ultra elite engineers/staff either retired or took job at lower pay instead of launching competitors to big tech.
Who is going to fund these smaller companies? Why do you think this isn't happening right now as we speak? Where are these companies made from these layoff discords in the past two or so years?
What makes you think it is not happening right now? Startups have to come from somewhere, and they don't have to say how they got together, in fact, it would be smart for them to keep that "layoff discord" hidden if it can keep the lawyers attention away.
And we here stories of "guy left big company to make his own startup" all the times. It is a common story for indie video game devs for instance.
I wonder if what we'll see instead is smaller startups replicating whole consumer value chains and replacing the whole industry at once.
On the B2B level, business is basically relationships, because you want to have a person that you trust who you can go to with any concerns about your multimillion-$ relationship. The relationships are the hard part, which is why the biggest department in any enterprise startup is sales, and requests from the salespeople usually dominate the feature roadmap.
But on a consumer level, most companies invest precisely zero in relationships, because it is far too expensive to build a relationship with 300M consumers who each pay you $50/year. The parts that matter are a.) building a brand image b.) delivering a "good enough" product and c.) doing it cheaply.
AI does extremely well at all 3 of those. It's great at generating fictitious but plausible-sounding content that makes you feel great about yourself. It's a passable approximation at explaining how things are made, enough that a small organization could probably deliver a good-enough product. And it replaces all of those pesky humans that require HR departments and legal employment contracts and sales relationships and customer service. A significant portion of the work in any industry is simply greasing the supply chain, making all those transactions that need to happen along the way happen. Vertically integrate and automate those and you could probably rebuild the industry in a much cheaper, more efficient way.
I think many people are thinking too small with AI, on the level of "how will this help me do a task?" or "how will this help my job?" or "how will this help improve my company by getting rid of jobs?" The real money is in "how will this revolutionize the industry by getting rid of all the other companies?" or even "how will this replace society so there is a fair playing field for new industries to get started?"
it completely depends on the elasticity and shape of the supply and demand curves. Some will choose to reduce headcount, some will choose to build more. Depends on the business and the nature of the industry.
This... It's weird how companies and people think adding something is the same as adding value.
Feature creep was an issue even before AI, now we see one worse feature being added after the next (does anyone even know all the touch controls of iOS or even macOS, as an example), with some expectation of that there will be money coming in because of it.
I benefit greatly from AI myself, but I use it to augment my capabilities, not as a slot machine for building things nobody will ever use, which seems to be the norm based on the products I use.
10x the work requires exponentially more communication and coordination. At least according to Brooks "Mythical Man-Month" thesis. There's more to creating a marketable product than churning out code. I'm not aware that AI really changes this.
I think this is a better analogy, particularly because it illustrates that the more demanding company might not need to be 10x more efficient. A company that opts to to be lighter/cheaper may lose to the competitor regardless, especially in winner-take-all or discerning markets. Maybe not always, and we're obviously speaking theoretically, but it's an interesting thought.
Most non-commoditized reasonable paying jobs are a product of information asymmetry. Anything that shrinks that asymmetrical surface is a threat to jobs. The company you are missing is the one that fires 99 of their engineers, hires 200 generally lower performing but far cheaper engineers and uses AI to accelerate their output.
Now whether or not that would actually work given the reality of the AI output is a question, but that's the general premise.
There is an assumption being made that more engineering leads to more company growth. It just isn’t true though. Sales sure. But doubling the engineering floor does not double sales. You can sell a bug filled product with an unbuilt roadmap still.
There's just one small problem. The AI costs money, and the premise is AI is cheaper than people. If you augment your 100 engineers with AI, they'll cost what, 10% more? That's +10 percent to your cost center, and would be considered very bad. There are tons of tools that make us more productive - think about it, how many of them did we have to beg demand or leave for? You assume they care to maximize productivity, which is your first mistake.
I don’t think anyone has figured how to do 10x the work yet as we don’t know what the work is. AI cannot (yet) define the work that needs to be done, for some definition of needs to be done, and we need humans to figure that out for now. Us humans cannot work fast enough so we still build the same stuff but with less people.
Overall US economic productivity has gone up about 6X in 75 years, which included computers and the internet. So we absorb these things, but gradually.
> There is no obvious point where the microcomputer or the Internet made any major difference
I think it lags. There was a sag between the mid-70s and mid-90s, which is when computers and the Internet were being rapidly developed but weren't really hitting the business mainstream. But late 90s through today it clearly ramped up a lot -- my bet is this is because of IT. It just takes a long time, it's not a sharp jump because not every business gets on board immediately and productively.
My guess is AI ends up the same. But then again, I'm not an economist, only a developer, so I just think I can tell anything about that productivity graph.
Are we looking at the same graph? What I see is that there was a huge spike at the onset of COVID, and then it flatlined and even corrected a bit, and in the end seems to have resumed the same trajectory it had before COVID.
If I understand correctly, the spike in 2020 was not any kind of jump in real productivity, either, it is an artifact of how the calculation is made -- there was a sudden drop in hours from low paid workers who typically have a lower than average contribution to productivity (as measured), so laying a bunch of them off skews the per-hour calculation. Then as things settled down companies re-hired, and then over-hired.
The covid productivity spike is because all the low productivity retail workers got laid off. Then as they came back to work productivity went back down
White collar work may have been phoning it in (literally and figuratively), but the drop is probably more likely due to "Labor Productivity (Output per Hour)" being measured using revenue from things like the social, entertainment, and travel industries taking a hit in customer spend (an empty restaurant has less output per hour than a busy restaurant, for example). It's not clear to me from that page/graph alone that an increase in stimulus spending is a cause, vs a correlation, for the drop.
The one who fires 90 engineers, just to see how things go with the 10, and then scales back up with cheaper off-shore talent if they can't keep up with demand.
Consider two reasonably similar competitors with 100 horses two years ago. Suppose AI really does accelerate development 10x times. One company fires 90 horses and buys 10 cars. The other company keeps all its horses.
Right. People don't understand the difference between labor and value. The job of a corporation is to extract as much surplus value from its workers as possible while making it as opaque as possible to individual workers the actual revenue they bring in.
It's very normal to think that the extra revenue growth you bring in should mean you have either 1) a larger share of the pie for every efficiency increase that grows revenue or 2) less expectation from you, as a worker, to labor as intensively since you are continuing to bring in surplus value that is being actively skimmed from.
>It's very normal to think that the extra revenue growth you bring in should mean you have either 1) a larger share of the pie for every efficiency increase that grows revenue or 2) less expectation from you, as a worker, to labor as intensively since you are continue to bring in surplus value that is being actively skimmed from.
It's also very normal to force businesses by law to either or both. But it's not happening these days.
Arguably, I would say this has been a 50+ year-long project starting in the Reagan/Thatcher era that Clinton/Blair fully realized by ushering in de-regulatory, labor-disempowering neoliberal policies as part of the left political flank. There's been a lot written about this:
I dislike the phrase neoliberal (its badly defined). I would argue that Blair (and Brown and Cameron etc.) was a lot worse than Thatcher because Thatcher did want competitive markets, whereas more recently its just been "business friendly" which is quite the opposite.
Whenever I read comments like these I feel like people are living in a completely different reality.
Most of the tech companies I have worked at are ridiculously easy and generous, I get to work with smart people, and the entire job can be done in a few hours per week. It's not like we're doing manual labor and digging ditches; half the people on this website are laptop-class workers who occasionally use the slightest bit of energy to write a few lines of code or participate in a meeting.
I show up to work and I often do extra because I like it.
The company isn't "extracting" anything from me except the extremely consentual labor I want to give back simply out of the love of the game. If you don't feel that way, the problem isn't the capitalist system it's probably your own life choices.
You’re reading negative valence in technical language that doesn’t mean it. ‘Extraction’ and ‘labor’ aren’t bad, they are technically descriptive. You extract fun and cuteness value from puppies and LeBron labors at basketball. Not an unhappy process in itself, though of course abusive versions exist.
That’s why your happy experience is not inconsistent with this description.
I completely agree and I think smart companies also understand this. There is still a lot of leverage employees have, especially in tech. Tech as "adult daycare" isn't a meme for no reason.
None of this changes the facts of the comment you're responding to. Every business must extract surplus value from their employee's work to be profitable. Nobody said every business has to be miserable, or not mask this extraction in fluffy terms. Doesn't mean it isn't happening though
How else would they stay in business if they did not make money off your skill/services? Unless you are a non-profit, why would you hire people just to break even or lose money?
The correct comparison isn’t him working by himself but with a similar sized cohort where distribution of revenue was fairer. Co-ops are a thing as are the productivity benefits of working with others.
Even if you don’t like the concept of co-ops it’s fairly easy to ask yourself if you would you be more or less motivated if you shared more in the success of a company rather than it being sucked up by shareholders?
You think a company can exist with just executive leadership?
Would love to see just one example of a billion+ valuation company with nothing but C-suite and that against the ratio of companies with 1000+ employees who are worth a billion+. The only two sub-dozen workers that come to mind for me are Theranos and Bernie Madoff and I'm not sure they are examples of success.
I think the 80/20 rule applies to employees correct?. 20% of employees actually do the work that is making money. So for every 8 sitting around doing nearly nothing there's 2 that are actually directly related to revenue.
At this point I wonder if those aren't simply marketing shills or the usual kind of "pro-business" press articles that always push ridiculously optimistic takes on new technologies.
Reminds me of the old press pieces along "with the new machines, the workers won't have to do this tedious and dangerous work anymore and will have so much more free time!"
What is missed is free time is just one good you can get. People like different "things". Sometimes that "thing" is free time. Sometimes that thing is more expensive food. Sometimes that thing is a bigger house. Sometimes that thing is... There are a lot of options. If you were willing to live like 1950 you could have the male working 20 hours a week and the female never leaving the home. What people forget is in 1950 having a black and white TV was unusual (though not uncommon) and color TV was still several years away. Your phone was a shared party line. You had one car which wasn't reliable by modern standards.
> If you were willing to live like 1950 you could have the male working 20 hours a week and the female never leaving the home. What people forget is in 1950 having a black and white TV was unusual (though not uncommon) and color TV was still several years away. Your phone was a shared party line. You had one car which wasn't reliable by modern standards.
Another way of looking at that is you could own a home and a modern car, and support your entire family and have the latest technology all on a single income.
TV wasn't worth having in 1950 for most people, if they even got reception at all, but by 1955 75% of homes had state of the art TVs. Phones improved over the decade as well, and pretty much everyone had at least one in their home (technically owned by AT&T). What people had seems primitive to us, but it was all new and constantly improving to the people at the time.
Today our experience causes us to expect that greed and enshittification will make everything both worse and more expensive, but back then people were excited about the future and about how technology would make their lives better while getting cheaper because that was their experience.
I think in this case, it was rather cynical. The "free time" the workers got was because they were unemployed. The workers who were still employed of course had no additional leisure time, the added productivity went to the employer.
Honest question: Where does the “AI booster” phrase come from? I only ever see it in online comment sections where certain people use it as a pejorative like this and everyone who uses it puts forth very similar ideas. Is this catchphrase something that comes out of a certain online community, political circle, or influencer?
I know you can find it if you Google it, but that’s my point: It’s the shibboleth that these people use to signal their alignment with an ideology, and it’s all the same thing.
I could have told you what those articles would have said before I began reading them because “AI booster” is a phrase associated with a very specific worldview. It gets dressed up prose about superiority and condescension for the “boosters” but it’s always the same thing.
It’s like many of the academic phrases that develop to signal alignment with an ideology and separate the insiders from the outsiders.
This is pedantic but that's not proper usage of 'shibboleth.' The basic conceit of an 'AI booster' is easily defined by the word 'booster' itself - it's someone who 'boosts' AI, likely without much actual subject matter expertise on the matter and the word 'booster' has a long etymology applied to various ideas: https://www.etymonline.com/word/booster
When most theorists talk about shibboleths and impenetrable academic writing, it's in reference to terms like 'deterritorialization' or 'intertextuality' which are not immediately evident to most lay people who aren't armed with an etymological background or who once stumbled into an advanced course on literary theory in pursuit of a creative writing degree
You could probably say that there's cultural capital associated with it but that's true of nearly any term describing an ideology or opinion. The very same AI boosters are guilty of coining things like 'orthogonality' or 'instrumentally convergent subgoals' which lean more towards being unnecessarily bombastic than descriptive (I guess 'intelligent but amoral' and 'self-preservation' aren't fancy enough for adoption) which loses the strongest argument for academic jargon like the one advanced by Butler: https://web.archive.org/web/20240630040227/https://www.nytim...
To me it seems like a term that users pessimistic about AI and the future apply to those with an optimistic opinion.
By speaking of boosting, they add an implication that the opinion isn't as genuine as their own and is perhaps motivated by some sort of financial benefit.
Do you believe than an industry that's investing 1 trillion per year in infrastructure, isn't also investing tens of billions in marketing, including pseudo-grassroots movements aka astro turfing? They would be total idiots not to, and they're not.
I prefer that we discuss the people and situations for what they are, not reduce them to catchy phrases that lump people together and discard any nuance.
If you don't have a cynical view about it, then why do you feel bad about being called a booster of it? Does it not describe your view? Or, if you are like not cynical but still like lukewarm, why do you feel targeted in the first place?
If everyone had the same opinions they do, but called it all "AI optimism", or whatever other term you prefer, would anything change for you?
This feeling you might have, where everyone is having a secret meeting without you, deciding on opinions and terms to operate against you, is understandable in this world, but you can't live forever in that solipsism. You need to be brave enough to have your own opinion against those that truly, passionately have their own. You need to argue and fight for what you believe in, not try to explain away or conspiratorialize your opponents. There is no one coordinated, arbitrary force you can point to that can explain why people don't share your worldview. You can just defend what you believe in, and always be ready to update your assumptions.
I've seen Moonshot pay creators on X for posting promotional messaging about Kimi K3.1 Pro, with the requirement not to disclose it as a paid sponsorship.
X then took action and suspended creator's accounts that were posting about Kimi K3.1 without the paid sponsorship tag, allegedly hitting some accounts that weren't paid by Moonshot.
In any case, I do not think that people are typically being paid to promote AI generally, rather than specific products.
To imply so on comments from well established HN users, as if it was unlikely they simply find value in AI and have a genuinely positive opinion, often seems misguided.
> This is pedantic but that's not proper usage of 'shibboleth.' The basic conceit of an 'AI booster' is easily defined by the word 'booster' itself - it's someone who 'boosts' AI,
It’s a shibboleth because it’s a highly specific phrase used by these people.
The fact that it’s self-explanatory doesn’t take away from that.
It’s a shibboleth in the way it’s used. I see the word “booster” applied more to AI than any other topic because it’s the shibboleth chosen to represent that specific set of ideologies about AI.
> It’s a shibboleth in the way it’s used. I see the word “booster” applied more to AI than any other topic because it’s the shibboleth chosen to represent that specific set of ideologies about AI.
It's not a shibboleth because "AI booster" is a fairly straightforward label to describe a group of people with certain characteristic. Which is proven because as I wrote "AI booster" above, mentally I was trying coin a descriptive label rather than use a predefined stock term. Personally, I have no fixed label for that group, and could have just as easily written "AI enthusiast" or "AI promoter."
IIRC, a shibboleth is a characteristic word or pronunciation that gives you away, often without you even realizing it. Understanding that there are people who support AI and others who oppose AI, and trying to refer to those groups, is not that.
> Which is proven because as I wrote "AI booster" above, mentally I was trying coin a descriptive label rather than use a predefined stock term.
AI booster is a common predefined term, not something you coined on the fly.
If you look higher up in this thread, the other arguments are that it’s a common predefined term with wide usage and understanding. You were repeating a phrase that you saw before and associated with a set of ideas.
I feel like you are trying to paint it as a shibboleth because you feel targeted by it or you dislike the ideology of people who use it but it's probably just a banal reaction to the "doomer" adjective that I have heard before it and is even more entrenched
It just makes sense that the bigger the agora in which a debate takes place, the more nuances will fade away until you can roughly sort people into the "X bad" and "X good" buckets. Is it desirable ? I don't know. Is it some group's lobbying/conspiracy ? I don't think so
I understand what it means. That’s not what I’m asking.
“AI booster” appears 100X more often than I ever saw “cryptocurrency booster” and the phrase is always associated with people who have a specific ideology about AI.
It speaks to a lack of understanding of the economy as a complex adaptive system.
Same idea with all these agents that are going to negotiate our insurance down, haggle with our cell phone providers, and find the best hotel deals.
Your counterparty will have its own agent, a larger budget for that agent, and a lot more skin in the game to ensure they retain their advantage (cell phone bills are 100% of the cell companies income but is only 0.5% of your spending). Good luck.
Cell phone provider is funny one here. Effectively most efficient thing to do is to get a deal from competitor and then accept winback offer from them. You can not even properly negotiate for deals very often but somehow those winbacks end up being near best deals. Last time I got negative deal for them. Meaning my mobile and wired internet ended up less than wired internet was before...
> Your counterparty will have its own agent, a larger budget for that agent, and a lot more skin in the game to ensure they retain their advantage
I did some work on negotiating agents. This doesn’t have to be a zero sum game. If you can craft a per-user package effectively (for many services) then both sides end up with a better deal, and competition in the market stops the provider entity from dictating terms.
A lot of systems are either adversarial or the v0 agents are relying on a sort of free rider / human good will that won't scale when 100M people have agent bots.
Apples iPhone call answering for unknown numbers was OK until all the overseas robocallers made their robot respond somewhat conversationally and hang on the line so that it vaguely looks like a real enough call. I can still tell its a robot since I am not senile yet, but I still have all the notifications about the call/transcription/voicemail/etc to tap.
People deploying agents to call restaurants for reservations works for like the first 10 calls that restaurant gets and now they are going to hang up on everyone's bots. Many websites already blocking agents. Some websites where agents were scraping for reservations/inventory are outright locking peoples accounts. Etc.
I’ve been baffled by how common these illusions about the labor market can be among people who have been isolated from the realities of the world in their career. I most commonly see it in juniors who never had jobs before graduating college, but I’ve heard it from adults who didn’t realize how much of their career fell into a very lucky, cushy track.
One of my strangest memories was a junior hire who decided after a few months that they didn’t like the work they were hired to do (which was exactly the job description, no surprises).
They informed their boss that the job wasn’t working out for them so the company would have to find some different work for them to do. There were a lot of difficult conversations trying to explain that they were hired to do a specific job that was needed, not to become part of a family where everyone belongs no matter what. This was a small company that didn’t have other open roles and wouldn’t have the budget to backfill even if they could move the person.
Yet it didn’t register. The person viewed their job as a jobs program, where they were chosen to join the company like a family and then it was the company’s job to find enjoyable work for them to do. They saw the paycheck as owed for their presence, and work produced for the company to be a nice bonus when it worked out.
It was a difficult situation until the person was finally let go because they could not accept the idea that their job was to do work in exchange for pay, and that the company wasn’t going to pay them if their work was not producing the necessary.
That’s the most extreme example but I’ve run into this a lot more in recent years. It got really bad again for the cohort of graduates who entered the job market post-COVID and thought getting jobs was always going to be this easy. I’ve been helping with resume review on and off and it’s stunning to see so many people who worked for 2-3 years post-COVID and can’t really tell you what work they did, but are so angry and shocked that their employers let them go.
> not to become part of a family where everyone belongs no matter what
I think this isn't helped by the fact that HR in alot of companies in the past decade or so has been going with this narrative. We are all a family together in this, stay late when needed, work extra/after-hours, etc.. Lots of places even have the expectation in this regard.
Of course this all goes away instantly when you get cold-emailed a layoff for example. I think its a systemic problem that originated some decade ago in an attempt to exploit people's judgment by creating self-doubt (ie: work more/harder under the premise that if you dont you are letting your "family" down), got perpetuated around alot, all job postings are worded like that nowadays too ("join our family") and it gives the newcomers to the job market the wrong expectation.
The worst offenders are all under IT companies as far as I can see, but because it's been going for so long it wouldn't surprise me if this is industry wide at this point for non-manual-labor job offers.
The "we're all a big family" narrative is a psychological ploy to try to exploit the need by most people to feel like they belong to some group. This need is deeply interwoven into our psyche. And as more traditional structures of society (extended family, churches, social clubs, etc) erode, it leaves a void that many cannot really understand. IMO, this particular message is almost never true.
Company "mission statements" and such are a similar tactic. Though I think it's a bit less cynical, because the company leadership sometimes (rarely IMO) believes in the mission to a minor degree.
I'm personally more concerned by the entitled goobers who think it's perfectly reasonable to feed the hardworking majority into a woodchipper if it multiplies the money in their brokerage account. Those are a structural problem, unlike kids who need to learn and will be taught.
I'm simply relating something I have observed through experience: the woodchipper goobers are excellent at victim blaming. If they wrangle you into that woodchipper, they will absolutely say it is your fault.
It's a strange critique to personify a structural problem in this way. It neither holds any individual accountable for their actions, nor points towards a better structure.
I think the market will teach them the lesson, because they have no political power with which to avoid learning it. Unlike the other goober I mentioned, who absolutely has power and is planning to use it to make sure the woodchipper stays fed!
Would be nice if we lived in a world where if you were getting "[fed] into a woodchipper" you could ... oh I don't know ... QUIT! At will employment. You're free to go and live your life how you see fit. Well, at least until we give the power back to the know-better do-good commissars and bureaucrats who then tell us everything we can and can't do for our entire existences.
"We have no work for you but you must work or you're lazy" is a ripe flavor of bullshit that really outsmells vague threats about how labor policy tweaks today will surely lead to the holodomor tomorrow.
Yeah, the kids are too naive about the job market.
On the other hand, it's not in anyone's interest to tell them the accurate version, right?
If the same kid instead spent their evenings working unpaid overtime for the company, because they would see it analogically to helping a family in need... the boss probably wouldn't tell them that they are making a mistake.
Heck, if someone just started telling those kids "the company owes you nothing, and you owe nothing to the company, the relationship is purely transactional, either side will be replaced when convenient", probably many people would object.
So yeah, the kids are sometimes confused by the confusing messages they get.
So basically the kid bought into exactly the corporate propaganda which has littered almost every workplace for the past two decades. I am not gonna blame the kid here but rather a culture of hyper-capitalism which is creating this illusion seemingly on purpose in order to get workers who are more loyal to their exploiter over their own material conditions as workers.
You are missing a critical third - people organize and via unions or the polls push back and require less work. There is no reason we aren’t working seven days a week beyond people fighting not to.
I'm very pro-union but I feel like at this point attempting to organize will just speed up the process of replacing as many people as possible with machines, not slow it down. It's too little, too late.
I don't understand what kind of society we're building right now. Anthropic and other AI labs at $20T valuations, the 1% monopolizing all the wealth and the rest of us out of a job and fighting for the scraps?
There seems to be some sort of magical thinking within the AI crowd at the moment, that it'll all work itself out in the end.
Software devs have been devastating industry after industry for the 30 or so years I've spent in my career. If you want to see the future for software devs look to the past and see what happened to everyone else ( good and bad ).
I don't think unions are the answer, at least not for knowledge workers. The knowledge of the workers is the bargaining power in that case and AI makes the knowledge worth way less than it was before.
You are 'pro' an abstract, unrealized union, which is not 'pro union' at all.
You have abdicated your obligation to align your actions with your stated beliefs.
Do not give up before trying. Do not believe the hype that your labour power doesn't matter. Even Amazon couldn't stop their most exploited workers from unionizing, is your company bigger that that? Is your position weaker than theirs?
Worse that can happen is that you started talking with your fellow workers and they aren't interested.
Just because SV is run by several millenarian cults doesn't mean you should take their delusions about neo slavery and feudalism (Thiel/Yarvin; your doomerism stems from here) or apotheosis (anthropic, EA) as inevitable. You should assume they believe their own hype, but not assume they understand material reality outside their insular social bubbles.
Thiel's last 3 or 4? attempts at building his techo-utopia collapsed before even starting; They have no motion outside of what they can purchase.
The position is weaker in the sense that an effective SWE union probably has to span national boundaries, and will likely be a short-term loss for SWEs in high-cost geographies.
> I'm very pro-union but I feel like at this point attempting to organize will just speed up the process of replacing as many people as possible with machines, not slow it down. It's too little, too late.
> I'm very pro-union but I feel like at this point attempting to organize will just speed up the process of replacing as many people as possible with machines, not slow it down. It's too little, too late.
Honestly, I think the only shot workers have is unionization plus really aggressive political action.
The problem is successful action will require setting aside pretty much all the culture war issues that are used to divide the working class, and I don't really see that happening. If anything culture-war polarization is getting worse. It will also require some formerly-privileged groups like software engineers to jettison their self-deceptive libertarian delusions, and properly understand their true role in society (software engineers should have unionized pre-2022 when we had more economic power, but economic comfort and those delusions prevented it).
The trouble with unionization is that it doesn't really address supply and demand, just bargaining. There are a dozen un-pulled levers available to bring aggregate labor supply and demand in balance. Overtime law, vacation policy, retirement policy, education funding, tax law (this is a big one -- taxes are obscenely lopsided on the labor/consumption economy vs the capital/investment economy), and so on. We need to pull those levers instead.
All those should absolutely be addressed (especially taxes), but organizing labor could be both a powerful way to push for those changes, as well as a force to maintain better labor/capital balance going forward.
> The trouble with unionization is that it doesn't really address supply and demand, just bargaining. There are a dozen un-pulled levers available to bring aggregate labor supply and demand in balance. Overtime law, vacation policy, retirement policy, education funding, tax law (this is a big one -- taxes are obscenely lopsided on the labor/consumption economy vs the capital/investment economy), and so on. We need to pull those levers instead.
All those other levers are political policy levers, correct? Besides bargaining, unions can also do political organization and lobbying.
Major state interventions that altered labor supply or rebalanced wealth (eg such as standardizing the 8-hour workday, establishing public pensions, or funding universal education) were won through waves of union mobilization and working-class political pressure.
It's not one or the other. Unionization is literally (figuratively) the hand that pulls those levers.
There is a reason nobody is willing to capitulate on the “culture war” positions they care about to “unify the working class.” People aren’t fungible economic automatons pursing material gain and they don’t believe policy is self-executing. If people have fundamental differences in how they view the world, what they value, what they think “good” is, they’re not going to capitulate on those in support of paper economic policies that may or may not work anyway.
> There is a reason nobody is willing to capitulate on the “culture war” positions they care about to “unify the working class.” People aren’t fungible economic automatons pursing material gain and they don’t believe policy is self-executing. If people have fundamental differences in how they view the world, what they value, what they think “good” is, they’re not going to capitulate on those in support of paper economic policies that may or may not work anyway.
I'm not saying capitulate, I'm saying set aside, as in table them to focus on more pressing issues.
Preserving some working class power in the face of the kind of automation being imagined now is going to take a large, unified political coalition that can prioritize effectively. Precisely because "nobody is willing to capitulate" on those issues, make them very useful in fracturing that coalition into non-existence or ineffectiveness.
If the "other side" doesn't put aside the culture war issues too, then "our side" will lose on them.
If you would explicitly state which culture war issues you are talking about, we might be able to have a productive discussion about it. Some issues are less important than others.
> If the "other side" doesn't put aside the culture war issues too, then "our side" will lose on them.
You can't have it all. Would you rather burn your energy to make little progress on them and lose on working class economic issues?
Call a truce, aim for a stalemate. Tolerate people from the "other side" to make more progress on other issues. AI has drastically changed the situation, and it's a massive fuck-up to continue to run the same strategy.
> If you would explicitly state which culture war issues you are talking about, we might be able to have a productive discussion about it. Some issues are less important than others.
The normal list: abortion, gun control, LGB and especially T stuff, etc.
You can’t “table” cultural issues because people’s view of what’s good and desirable arises out of their culture. That makes it extremely difficult for people with divergent cultural views to cooperate on shared economic goals. Viewing this as some top-down effort to “fracture the coalition” overlooks that the coalition doesn’t exist in the first place. If you look all over the world, people unify over ethnicity, culture, and ideology—usually in that order.
> You can’t “table” cultural issues because people’s view of what’s good and desirable arises out of their culture. That makes it extremely difficult for people with divergent cultural views to cooperate on shared economic goals.
What I'm saying is people need to learn to be grown ups and work with people who are different than them on common issues.
> Viewing this as some top-down effort to “fracture the coalition” overlooks that the coalition doesn’t exist in the first place. If you look all over the world, people unify over ethnicity, culture, and ideology—usually in that order.
It's not a top down effort, but I wouldn't but it past oligarchical economic interests to fan those flames a bit. These issues fracture the working class coalition nonetheless.
And it's not like this is some kind of impossible ask: the Democrats of the 1980s (say) did a far better job of it than the Democrats of the 2020s.
It seemed to me that the Democrats of the 1980s achieved almost nothing to stop or to slow down the biggest threats to working-class economic power of the 1980s, namely low-skilled immigration and offshoring (then called "globalization").
> The Democrats (in the US) of the 1980s did very little to stop or to slow down the big threats to working-class economic power of the 1980s, namely automation and offshoring (then called "globalization").
Wasn't that the 90s? IIRC, globalization was mainly a Republican big-business project, and the Democrats of that era were the biggest source of opposition, and it only got passed because of some defections.
Also globalization was sold as a bill of goods. IIRC, it was claimed (probably by many economists) that it would increase domestic manufacturing, but when that was shown to be a lie, there was no course-correction, and the sales pitch was changed to "but you got cheap stuff." I hope people have learned their lesson from that, to not trust economists.
But that wasn't really my point. My point was they were better able to handle divisive culture war issues, so you had lots of Democratic representatives from "red" states, even with popular Republican presidents. Today, the only way Democrats can hope to win those seats is if they're handed them through abject Republican incompetence.
My recollection is that "globalization" was a buzzword in the early 1980s although there is a possibility I am not accurately assigning my memories to the right range of years.
It was often paired with another term which I cannot confidently recall, maybe "the information economy".
Although Chinese imports didn't become an important factor in the US economy till the 1990s, Deng's rise to power happened in 1978 and his famous "Southern Tour" happened in 1982. Both events committed China to a course of economic modernization, welcoming Western investment and special economic zones designed to boost exports, which the US media did not fail to notice.
Trump was the first president in the modern era to try to stem the flow of factory jobs overseas. Clinton for example was all-in on globalization.
> What I'm saying is people need to learn to be grown ups and work with people who are different than them on common issues
This is like saying “people should grow up and learn to solve their disagreements without war.” I’d argue that view is the one that exhibits a childish utopianism. It overlooks what actually drives the formation of coalitions.
This is my prediction. The extractors will continue to extract until it becomes untenable. Then a lot of people will die and we will find a new balance. This cycle is basically human history in a nutshell.
Do those calling for unionization forget how the railroads dealt with workers trying to unionize? Ever heard of the Pinkertons.
Large companies have been spending millions in propaganda and attacks to ensure that people in the US think that unions are pure communism. By the time we get enough people to make a difference there may be T100's stepping on our bones at this point.
The problem is everyone's financial cliff is different. If I lost my job today, I have enough saved up to last me X months. Yours is Y, someone else's is Z. I'm going to go hungry at the end of X plus whatever. But since you're going to go hungry at Y plus whatever, we aren't forced to agree on what the right time to stop and stand up is. Z over there has rich parents so they've got longer. Nevermind poor W over there who doesn't, and has shorter.
What a convenient little trap we've set for ourselves. If everyone's money was set to run out at the same time, there'd be problems.
There were church mandated holidays in the Middle Ages. Lots of them.
The current pope chose his name to signal his continuation of Pope Leo XVIII's teachings on worker's rights, particularly the papal encylical Rerum Novarum: https://en.wikipedia.org/wiki/Rerum_novarum
During the industrial revolution many people treated their workers better because of their religious belief - the Quaker industrialists of Britain are a prominent example.
Only replying to you since everyone else is making the same points.
I follow a religion with similar edicts (not Christian). I also understand the contributions made by Christianity in this area. But the fact of the matter is that religion was ultimately unable to force actual widespread change on the ground, especially not in secularized European states.
At the end of the day, the only true religion that industrialists and capitalists understand is money. Only unions were able to speak their language at a very large scale, and this is the language they ultimately had to listen to :)
Both, actually! Nonconformist Christianity was one of several key influences on the labour movement ideologically speaking. At least in the UK the Labour Party is often said to owe more to Methodism than Marx for its original ideological foundations. There's a long history of this, in the aftermath of the English Civil War a group called the Diggers more or less independently invented agrarian socialism out of the New Testament, and a great deal of calls for humane conditions for workers over the centuries were motivated explicitly by Christian teachings. The New Testament is pretty clear in its condemnation of repressing workers, James 5 for example:
> Go to now, ye rich men, weep and howl for your miseries that shall come upon you. Your riches are corrupted, and your garments are motheaten. Your gold and silver is cankered; and the rust of them shall be a witness against you, and shall eat your flesh as it were fire. Ye have heaped treasure together for the last days. Behold, the hire of the labourers who have reaped down your fields, which is of you kept back by fraud, crieth: and the cries of them which have reaped are entered into the ears of the Lord of sabaoth.
This isn't an exclusive or situation, nonconformist religious influence helped create the ideological environment that allowed unions to exist in the first place.
The "cushiness" of jobs such as fewer work hours, more benefits, etc. preceded the formation of unions and then slowed down (overall) once unions became popular.
It has little to do with union pressure and a LOT to do with economics, in particular the demand for labor. Jobs get nice when employers are competing over the same pool of employees.
Unions are nice for those in them and bad for everyone else. They are (quite literally) cartels, in an economic sense.
I don't know enough about the US, but for Europe this is 100% false! The first unions date back to the 18th century and they were an important part of the worker's movement in the 19th century. They absolutely played a crucial part in improving worker conditions during the industrial revolution and beyond.
> Jobs get nice when employers are competing over the same pool of employees.
Maybe in high-paying jobs that require high qualifications. For most jobs, especially the low-wage sector, this is absolutely not the case. There is no competition between employees if applicants are desperate and need to take any job they can.
I don't have time to read these papers. What I can say for sure is that your first sentence
> The "cushiness" of jobs such as fewer work hours, more benefits, etc. preceded the formation of unions and then slowed down (overall) once unions became popular.
is demonstrably false.
Sure, unions are not the only reason for better working conditions, but they certainly played a very important role.
The legal restrictions on the max. number of hours people are allowed to work per day or week are only in place because of the worker's movement and the power of unions. Companies certainly didn't lobby for these laws. In fact, certain industries are constantly trying to soften or undermine it.
BTW, in my country, unions are part of the annual negotations between employers and employees ("Kollektivvertrag"), which most notably includes the minimum wages and annual pay raises.
If you don't have time to read the papers, you could at least hand them to your favorite LLM and ask them to summarize the strength of evidence for broader macroeconomic factors vs. unions as the primary drivers of work hours, benefits, pay, unemployment etc.
Yeah not sure you’re right on this one. I’m talking strictly USA here, as that is where I live and my anecdotal experience.
The first official Union was formed in 1794, for journeyman cordwainers, I believe in New York.
The national labor union was formed in 1866 and is the first official call to congress to regulate to an 8-hour workday.
The first example I know of on a national scale that voluntarily promoted an 8 hour day, 40 hour work week was Ford in 1914 and realistically, that is exactly what you stated, an attempt to undercut other automakers / factories of the time for labor. Not because he was being altruistic.
I’m sorry you’ve had bad experiences with unions. I am certainly biased, because I grew up in a Union household and have been apart of them myself, but I’ve only had good experiences. Yeah, there are some bad eggs, admittedly, but I think collectively they have benefited the working person.
> I'm sorry you've had bad experiences with unions
I've actually had personally great experiences with unions! But I try to not let my anecdotal data color my views too much.
I think this is part of the reasons unions are insidious - their benefits are quite visible and concentrated to those who are in them, and their costs are spread out and invisible (until you put an empirical lens on them at scale).
Society has settled on certain norms that shame certain behavior (6 or 7 day work weeks for example) but I’m sure unions are more agile in demanding certain things for groups that need more than those norms.
I am with you on the competitive workforce groups though. Large corporations don’t need to feed you three meals a day and bus you in from hours away but they do to entice workers. There are some economics at play that probably are more effective for certain groups of workers. The less skilled and more replaceable the job though, the more a business can take advantage because it’s a race to the bottom.
Without laws and unions, I fully believe some industries would still have slaves, horrible unsafe conditions, etc. Laws are probably more vital than unions IMHO
> Jobs get nice when employers are competing over the same pool of employees.
During an all hands where quarterly earnings were presented for a very large American company, the live audience were obviously employees.
The CFO came on stage and was very enthusiastic about the labor market "loosening". Employees were clapping. Unsure if because they didn't understand the meaning of that statement or because just to appear supportive of leadership.
I was thinking: they're talking about us, turkeys.
How do Employees advocate for themselves if not Unions? I've never heard any anti-union person give a good answers other than 'just trust the employer to be a good human' or 'wait until it gets bad enough that the market sorts it out' which is just laughable.
> How do Employees advocate for themselves if not Unions?
The same way you advocate for getting good bananas, good streaming services, or good laptops.
You only make voluntary transactions you want to make. If you don't like what a job is offering, take another job or start your own business.
If you don't like ANY of the bananas being produced by ANYONE, this probably isn't because all banana producers are moustache twirling villains. It probably has something to do with the underlying realities of banana production, supply, and demand.
I think you’re not seeing the whole picture about pushing back: It doesn’t happen in a vacuum. People have options to organize and push back but it doesn’t automagically make their demands possible. Companies also have options to push back, move the work to other locations, withdraw from markets that aren’t conducive to work, and now, look at automations.
Much of this unionization conversation has lot the plot about how unionization is about collective bargaining, but you still have to bargain. Both sides have options and you can’t come to the bargaining table with a belief that you can get whatever you want by threatening to strike or quit. There are limits to this and they’re decided by markets and technology.
and politics. The last wave of worker bargaining wins happened in the wake of the Great Depression when the last of the capitalist prosperity gospel promises were defaulted upon and the electorate finally became willing to consider alternatives other than giving rich people everything they wanted all of the time and hoping it would trickle down.
I'm hoping we don't need another Great Depression this time, a collapsing economy brought about by mistaken belief by those on top that they should have the right to keep it all... but that might be exactly what we need. We'll see.
Politics also doesn’t happen in a vacuum. You treat organization and politics like they’re neat tools that the populace only needs to wake up and discover, after which they get everything they want.
The Great Depression politics example is useless when we’re discussing AI knowledge work in a global economy. People in one country could organize to try to force companies to give them more money for less work that they can do with AI. They could try. It might work for a while. Then companies in every other country would jump on the opportunity to use the same AI to do the same work without those restrictions.
"Growth mentality" is only convincing if you're actually delivering the growth, though. If your threat is that demanding more will mean our billionaires have a lower number than their billionaires... your offer is not compelling. Take a hike.
I wasn’t making an “offer” either. I don’t know what you’re talking about.
I’m stating simple facts about how markets work. You could try to regulate to extremes that force companies (or your billionaire villains) to give everyone their money but without an actual sustaining market dynamic then it’s not going to get very far.
It came from you making the capitalist prosperity gospel / growth mentality argument:
> companies in every other country would jump on the opportunity to use the same AI to do the same work without those restrictions
"If you don't race to the bottom on labor, someone else will!" Two problems with that: one, the government can largely regulate access to its market -- the ability to do labor arbitrage is a policy choice not a law of physics -- and two, to the extent that the government can't regulate it because there is a finite appetite for protectionism, people still care about the growth delivered to them not to billionaire brokerage accounts, so you still have to normalize the growth to what you actually pay people. "Just let the market settle into a natural equilibrium of not paying people" isn't actually the win button that you think it is.
> Two problems with that: one, the government can largely regulate access to its market -- the ability to do labor arbitrage is a policy choice not a law of physics
We’re talking about work that AI can accelerate, which is knowledge work that happens on computers (right now)
You’re still not seeing how portable this is. You can try to regulate this to death and force companies in your country to pay high prices for things that are cheap now, then add tariffs or import restrictions to force consumers to pay high prices for things that are cheap now, but this doesn’t matter.
It’s a global economy. The world will be moving on with or without you. You can try to protect chosen people and industries for a while, but it is not going to save you from the inevitable.
It would be like trying to use the force of law to stop farmers from using powered farm equipment because it reduces farming jobs. Do you think your current country would be doing great if they had done that 200 years ago in the interest of protecting farm worker jobs? We wouldn’t be having this conversation right now because you’d be busy doing farming work so you could eat this winter. But your job would be protected!
The industrial revolution could have done a much better job taxing and regulating industrialists to reduce squalor. "They will outgrow us if we don't squalor-maxx" was a dogshit argument then and it's a dogshit argument now. But no, the industrialists were so dead set on ignoring social goals to maximize their own bank accounts that after growing to command their own country, and then the world, they eventually set eyes on each other and burned it all down. That was the point when labor was able to successfully demand and receive concessions which turned out to not be impossible after all! Even in the bombed out shell of a country that was left after the destruction wrought by the insatiable greed of the industrialists whose growth was supposed to be the only thing that mattered.
This time around, everyone has nukes and drones and AI bioweapons, so maybe we could consider the social goals first, and see if we can't grow on that dimension before we go through another episode of worldwide capital incineration? Just a thought.
That's what German unions did in the 90s. In the end it only undermines productivity of the national economy while competitor countries eat your lunch.
I'm pro unions and come from a social county. I'm self employed though, if I start relaxing, my "competitors" outrun me instantly. Bye bye assignments. This seems like primary school level reasoning to me, I also don't understand the premise of the article.
What stops me from working more than others is being smarter than my competitors right now. Not being smarter in any objective, fixed way. To keep it that way I must constantly evolved, learn.
Yes, micromanaging AIs is more intense for me than coding, and feels more tiring, I hope that (or I) will change, somehow.
I started asking for specific example/answers each time a conversation stopped before of the end of the task. I got the sense that frequent checkpointing is baked into Opus 5.5. I then added some language that amounts to “carry this through to completion only stopping if you’re truly blocked” and it really helped lengthen the loop
I think even apart from unionizing, if all/ the employees at company aren't AI jockeying, more work isn't necessarily demanded. I realize you have to have relatively decent management too. I consider myself extremely lucky to work for a company that hasn't been beating the drum of AI maximalism and our work load in general has stabalized.
Obviously things can change but in general I feel AI jockeying is the exact opposite of showing solidarity with the people you work with.
Somebody should do a economic analysis of how technology affects unions. I bet the availability of automation has a lot more to do with the downfall of the unions than the political changes that are commonly described to that phenomenon. Similarly, I bet the availability of AI will have a devastating effect on any white collar unions.
"My boss would never dream asking I crank more because I now have AI help."
But... You're cranking more right? I mean, she may give you the courtesy of not asking for more as you (and we collectively) figure this out, but it's getting quite clear that you can crank more, so why would she not expect it?
Sure there may be a union, but if you're the only person not cranking out more with AI, there can still be consequences.
you're fortunate; the power of unions has been decimated since the 80s, and huge efforts have been made by companies (with the support of the government) to prevent new industries from unionizing
Once companies can just outsource anything from manufacturing to services the only categories that can really unionize with success are those that can de facto block everything at once (port workers are an example).
Unions still work? They exist and do real work for their members. They can also coordinate and cooperate which is, I think, how some unions managed to put pressure on Israel. Was it maybe Italian dock workers?
That globalism has made unions not-work is just Capitalist Inevitabilism. Yeah okay I’ll give up when the billionaires have fully autonomous robot brigades. But not over globalism.
That was a rhetorical question kind of response to you saying that they don’t work, except dock/port workers. By extension I have no trouble believing that those dock worker unions work.
Apparently, minute hands were added to clocks so factory owners could extract more work from their employees.
Medieval people used to own more property (though not technologically equivalent of course) and "owned" more of their time than the average person today.
----
As for the title of this post, I can testify to that: As a solo coder, AI has actually increased the amount of work I do - but in a good way!
Even on non vibe-coded stuff, AI reviews keep helping me finding bugs and inconsistencies to fix that would have taken ages to even notice on my own
but more than that, whenever I have the faintest idea about anything, I can immediately put it to practice and explore it further or move on to other ideas.
Even when I'm outdoors away from my computer doing some normal life stuff in meatspace, I often think of stuff and pull out my phone and open the ChatGPT app to remote into my computer's Codex.
> Medieval people used to own more property (though not technologically equivalent of course) and "owned" more of their time than the average person today.
That's a pretty serious misconception unless by "medieval people" you mean the nobility. Common peasants had some hereditary title to the land they worked but the land itself was owned by the lord and the king. They were more like hereditary tenants but much worse off contractually than our modern interpretation of a tenant would suggest since they (and their progeny) were often obligated to remain on their land and work it without the option to go elsewhere. They paid their rent in labor and produce or sometimes money if they had any. They had parcels of land that are large by modern standards because they were farmers and they had to sustain themselves on what they grew on their land as well as give a portion of what they produced to the lord that actually owned the land.
They may have "owned" more of their time in that they weren't employed in the modern sense but they still had to put in as much work as it would take to feed themselves and pay their rent. They absolutely had to perform the work their lord demanded of them which might include additional agricultural labor on other plots of land or threshing grain.
It really wasn't a better system. Agricultural labor is gruelling without modern equipment. It's even worse when your life depends on it. Not being allowed to choose where to live or choose your occupation and having that apply to your children and your children's children is worse.
Imagine if you were permanently contracted with your landlord. You could never move to another apartment and your children were obligated to live in the same apartment. On top of that you had to do the maintenance work for the building whenever the landlord asked you to and you still had to pay him rent on top of that. You'd still be slightly better off than a peasant because you wouldn't have to be a subsistence farmer.
I probably didn't say it right, but watch that video, Monty Python's Terry Jones explains it better than I can, agreeing with the points of other comments here.
Unions do not exist in a vacuum. They have to operate in the same fundamental power structures as everyone else.
If union A is like "work less let AI take the slack" and union B is like "work the same, use AI to achieve more, let AI help us win" I'm joining union B every time.
Unions also require the right structural conditions.
For example waste management generally has very strong unions in the developed world because 1) capital intensity 2) operations immediately seize on labour withdrawal 3) consequences of strikes are in a golden zone of painful but not existential so are tolerated 4) extract a relatively small proportion of local gov funds, so don’t create that much powerful opposition 5) labour force is locally concentrated and highly socialised.
Those pillars of power don’t rely much on any particular legal or political system.
We have our own small boutique software type of business, so we aren't dealing with "employers", but all of us have never worked more, and for less (per hour) pay. Customers simply have higher expectations now, and we have to compete globally with basically an entire planet willing to vibe-code things, along with internal pressure at clients of "we can just do this ourselves".
So that means shipping higher quality, faster, with more features, and for a lower end price.
> and we have to compete globally with basically an entire planet
This is the part missing from many of the hot takes about regulating AI: The type of work that AI can do is the most portable globally. You can try to regulate AI or force companies to pay more for AI-assisted labor, but that only works within your borders. A hundred other companies in other countries will be more than happy to come do the work remotely. You can try to regulate that, too, but then the entire products will be built outside of the country.
> What kind of illusion must one be under to believe that employers would ever pay you the same amount of money to do less work?
Because supposedly, the new tooling (in this case, AI) allows them to produce the same or greater output despite doing less work. Workers are under the perennial illusion that they'll ever benefit from productivity gains, but that hasn't been the case since the 1970s.
> Workers are under the perennial illusion that they'll ever benefit from productivity gains, but that hasn't been the case since the 1970s.
Agreed, except that workers have never benefited from productivity gains, those always accrue to the employer.
The wage growth before the 1970s was mostly from the postwar boom, before jobs started to be moved out of the country. After that there was downward pressure on wages.
In The Conquest of Bread Kropotkin talks about the massive productivity gains of Europe in the age of steam and steel, and uses numbers to show how there's enough beef and mutton and bread and chicken to go around and enough trains to get it to everyone who needs it, such that we can look forward to a future where a worker gets much more for much less work. And we have hit that in some respects: a chicken is nearly free now compared to its historic prices, and the 40 hour work week was not widely established when that book came out. So we work less and get more now, but we don't work a few hours a week in exchange for a life free of any economic insecurity, as has always been on the horizon for technology optimists.
The question is not whether working hours have decreased in history, but whether this has been a direct result of productivity increases as opposed to due to other factors, not just due to increases to productivity. I don't think most commenters here address that at all.
Not an economist but I've heard the argument that the 5 day 40 hour work week was only possible by the increased productivity that came from the industrial revolution.
My understanding of the argument is that it doesn't deny that unions and strikes weren't instrumental but that it was only economically sustainable because of the increased productivity.
Since most of us weren't still working 19th century style 14 hour shifts in factories, or toiling with mostly manual labor at the farm, but had like 9 to 5s / 8 hour work weeks it was.
A situation where there’s one seller (USA) in a world of buyers (Europe, Japan) is by definition an abberation.
It may have been good (for the US), but that definitely doesn’t make it normal, and unfortunately I don’t see how such a situation occurs again without similar circumstances. I’d love to be wrong, though.
Yes, with the formation of unions. The owners realized that they needed to placate the workers because damages were worse for business. So workers won a 5 day work week, 40 hours, and many more benefits.
I agree, but that is due to social struggle not just because more efficient means of production got invented, which is also what I think is the point and the reason I asked the question.
1970s? More like the 1800s with the industrial revolution. Workers have almost never seen increased pay from more efficient production - it's the business owners and investors who benefit every time.
People can argue that the workers benefit from buying cheaper stuff, but that's on the other side of the equation.
The real question is: what's the rate of growth for capital returns vs labor returns? If capital returns are accelerating, over 50-100 years, yeah, we will still end up in techno feudalism.
The point is work is exploding. Mgmt. until AI could be reasoned with. Trade-offs between different interests and stakeholders were negotiated like investing in features now vs. development velocity vs. long term architecture vs. bug fixes vs. production incidents.
Now with AI and the flood of CVEs, management anxiety makes discussions much harder. Throw in unrealistic expectations about productivity - derived from venture money fueled AI POCs in greenfield hotel california and stress goes up.
The CVE’s drive work, the introduction of AI drives work, the old ci/cd is creaking under the load and architecture really needs a rethink. Add to that the genius manager pronouncing that very soon nobody will be coding fueling the dissonance between what is needed from me today and where we are supposedly heading increasing motivation - not.
A rational employer is interested in how much value they can get, not how hard you work. In most cases, in the short term, an increase in efficiency means they can get more value from you working the same amount of the same pay.
Is the increased value coming from you? Or from the tool they purchased to improve your efficiency?
A dump truck driver doesn't necessarily get paid more than a taxi driver, just because he's hauling more material. It's all about the value you personally bring to the table, not the value of the tools you're using, unless it's a tool unique to you.
Well they probably do, as driving a dump truck takes more training and skill than driving a taxi. So it's a combination of the value you bring as well as the supply of other people who can provide that same value.
The problem is how 'hard work' is defined. Does staying after hours count? Not automatically, but managers everywhere use it as a proxy ( whether it is dumb position or not is an argument in itself ). Does carefully considering your options count? Does being anal when it comes to minute detail count? All of these are small dimensions of hard work, but we throw those vague terms willy nilly, because we hope the audience will accept our 'used key term values' within our message to them.
Keep in mind that developers at Amazon were pressured this year to burn through as many tokens as possible because their “rational” employer assumed that productivity equals tokens consumed. I experienced similar situations on my last project.
The abstract concept of value you mention is part of the same vocabulary managers use, and I believe they often end up confusing even themselves with it.
They are just as rational as any other human being.
I wait for the day when the employer, the management, the business analyst, and the customer can get to state expected behaviour. Nevermind using AI. On all my projects programme was never a problem, it was non-competent people going "Make it good, if it makes big bucks and line go up, don't even bother with that. Make what good? Everything.".
If you think LLMs will make irrational people go "Ahhh I get it, I can't throw money and yelling at a shower idea and be the next Xooglebookflix" you will be surprised.
An increase in efficiency in and of itself, maybe, but there is no such thing. The increase in efficiency has largely come from deskilling the job. Deskilling the job makes the employees time less valuable, because they can be substituted for another more easily.
[edit: it's important to mention that this is compatible with programmer ego trips. I consider myself a reasonably good programmer and I've been doing it for a long time, but AI has made me far more eligible for jobs in languages that I have never used on platforms that I have never worked with.]
But another factor is that just because you can do work faster doesn't mean there's more work to do, so if you were hitting all your deadlines before with two employees, now you can fire one. Which one that will be is the result of a reverse wage auction.
> Deskilling the job makes the employees time less valuable, because they can be substituted for another more easily.
That's not that self evident yet. If AI is a multiplier it might be the other way around i.e. can a $300k engineer + unlimited tokens generate more a $200k in extra value compared to a $100k engineer? That should be much easier to achieve with AI than without it? The more you spend on tokens the more it makes sense to hire the most highly skilled people you can find since their salaries are no longer that significant.
I don't think the dynamics are similar to what they are e.g. during the industrial revolution since AI is more than capable of clicking the submit button on its own.
The issue is that wages are based on supply and demand and rarely directly related to how much value you generate.
We had to riot and sometimes people were killed rioting to secure the weekend.
If things are to rebalance again and give some of the productivity gains back to the rest of us, we should be working far less and I hope it comes with less of a fight than it took for us to get 2 days a week off.
Yeah, I also don't get why people expect to work 2 hrs instead of 8 and receive same salary.
Once you go from horse to tractor the fields increase but work hours doesn't decrease. And if you still use a horse, you are outperformed by the tractor guys who will sell stuff at less the price that you can sustain.
Maybe so, but the deliverables expected of a full time worker tend to correspond roughly to some estimate of a typical worker's productivity. Or people are hired to perform particular kinds of duties, and expected to be reasonably productive in those duties given the time and resources allocated to them.
nope. it basically says "if we haven't fired you, you're good." (at-will)
it's kinda nice because you can manipulate your output and do less work. so long as my employer doesn't know my sausage is made, i'm free to add sawdust ;)
it's this way in the US too, if people would read the actual contract for a typical office white collar job 40hrs/week is in there along with whether or not you get paid overtime. Most IT workers do not get paid overtime but some do. In some US jobs paid overtime is required by law and in others it is not. Also, different US states have different labor laws too so there's multiple governments and laws involved plus the whatever the employee and company have negotiated between themselves. I can't think of a single labor law that applies to every single working person in the US no matter what they're paid for. [edit: except taxes]
Because that's what they observe the executive management performing for multiple employers simultaneously and the definitional difference between salary and hourly wage.
> Realistically there were only ever two possible outcomes: you do more work for the same amount of money, or you’re one of the ones who were laid off and your work was transferred to the former.
Workers need an answer to this problem and we shouldn't wait until all the deleterious effects stack up to make the demand.
I feel like this conflates work with productivity. One can be working more than ever and still be unproductive.
In my opinion what people think is that they will be more productive for the same amount of work, not have more work than previously and still only be approximately at the same level of productivity.
What AI has done to many is create more busywork than it solves in many cases, rather than making me more efficient for the same amount of working hours.
It’s been deployed widely enough that it’s becoming apparent where its lines are from genuine tool that boosts productivity to bolted on and makes things the same or worse, the pipeline for the work simply looking different
Perhaps is not correctly phrased, but the way I see it:
Because these tools are now available whoever that is up in the management chain (possibly someone that has no idea about code or technology) is now entitled to request the undeliverable in the same amount of time.
I believe a couple of scenarios might unfold in the near mid-term future:
* Programming becomes now a task that anyone can do (assuming models outpace even the highest standards when it comes to certain software qualities). Under this premise wages go to the floor, now even grandma can do it.
* Companies still demand competent people as models plateau and do not offer outstanding software qualities yet the output is good enough that programmers should be able to do way way more than before. The job should then be high-paid but turns out highly-stressful.
* For some years eveything goes "just fine" under the vibe code party until competent people at major companies realize that have disastrous products and basically unfixable. Models start to decay. Then old school competent developers become again highly required and only a minority of "skilled-previous-generation" programmers that decided to keep doing things the old way get desperately required.
Obviously no-one has the crystal ball and anything can happen...
Yeah, everybody wants freedom and flexibility from self-employment, but that's generally speaking not a realistic attainable goal. The most defensible small businesses are ones that are sweaty i.e. require a lot hard work and hours others just wont justify doing, and that's an extremely important facet because without employees its much easier for a business to catastrophically collapse. Pretty much by definition anything that requires low effort is easy to clone and thus become saturated.
Small business self-employment doesn’t care that much about saturation, I think, only the supply-demand relationship. See everything from Fiverr to coffee shops.
For years, decades even, employers were straight up offended by the idea of paying someone work 4 days instead of 5 days, while paying them the same. It made zero economical sense to them.
Then companies started testing it out, and have noticed that things like sick leaves plummeted, and that the employees produced the same amount, sometimes even more.
I never really understood that and i think it goes back to factory jobs where time was directly tied to widgets produced which was directly tied to sales. If i had a company and the bank account was increasing at an acceptable rate and my customers were happy with my product/service then i wouldn't care how many hours employees put in. As long as customer satisfaction is high and bank account balance keeps going up then what else matters from a business perspective?
> The only way it will result in less work and the same or better quality of life is if you’re self-employed or are employed by a compassionate human. Sadly that does not apply to the majority of people.
Even that is unlikely. If you are self-employed in any remotely competitive industry, everyone is continuing to work the same amount and AI is boosting their productivity. If you just start slacking off, your product will become inferior.
This then extends to the employment situation as well.
Only way you can get less work is if you (or your employer) are in a monopoly with a strong moat.
Because for salaried knowledge work, the premise is that your impact or output and not the hours worked is the basis for your compensation.
Suppose someone (pre AI) wrote 100k lines in 50 hrs/week of bad code but someone else wrote 10k good lines working 40 hours. In theory the former person would not be the high performer.
I agree many places don’t work that way, but also many do - so it’s maybe a hard discussion to have broadly and not scoped to the individuals limited experience.
I don’t think this was ever the idea. It’s always been pitched as a productivity improvement tool but AI leadership comms has clearly been a complete clusterf*k.
The idea is that each individual employee does more with less and therefore the price of the end product comes down, raising the standard of living for everybody.
Once a total economic conversion happens, which might take a decade, individual employees can definitely upskill and jump revenue bands. Electric utility workers get paid 10x more than the lamplighters of yore, but it’s just a different job.
> If the volume of work goes up, its cost goes down. Supply and demand.
What?! Good to know. When contractor comes to do the kitchen, I will make them do the garden as well and will reduce the money I’d pay them, because supply and demand!
> Unless you think you should pay more now for more content
We do already! You realize you’re paying by the token? More token -> bigger bill!
> The only way it will result in less work and the same or better quality of life is if you’re self-employed or are employed by a compassionate human.
May we take this as an opportunity to recognize what a bad deal we get from most employers and explore better ways to organize labor. It's never been more possible. UBI is unlikely to happen and uncertain it would even be beneficial. If AI can lead to a better world, this seems like the most promising path.
What can make sense, and what has been my personal experience, is that I have shifted my most draining work to the AI, so I can actually get more useful work done without getting exhausted. Just like 10 hours mining sitting inside of equipment is better than 10 hours with a pickaxe and a shovel, even if we ignore productivity
>What kind of illusion must one be under to believe that employers would ever pay you the same amount of money to do less work?
There's the historical experience with early/mid 20th century unions, the establishment of the 8 hour workweek and worker rights, even Henry Ford's wage increases.
Not necessarily that they'd do it out of their own good heart, of course.
This is the goal of every productivity gain from the worker's side, and owners failing to deliver for long enough inevitably results in revolution (usually after some demand-destroying crisis).
What you call an illusion is reality, and the notion that it's an illusion is absurd.
I am surprised people don't look at this like hiring a plumber. When you hire a company to do something for you if they are doing less work and/or the work is easy enough you can do it yourself, you would not want to pay someone a lot to do the work.
It does seem quaint in retrospect. Its not dissimilar to the the cost of capital will go to zero from certain parts of the finance world during COVID era.
What does compassion have to do with it? If the industry requires more output or your business fails what options do you have?
Also, as someone who has been self employed I can tell you that you're working 2x-3x as much as someone who is employed and you have to pay more in taxes.
Even in the first paragraph the author goes into some nuance about what they mean and it's not at all what you're saying? Years ago, I'd work on one or two things at a time. Now I do maybe 10+ things in parallel. I work more hours, "let me just finish up this one thing" I'll say, and I look at the clock and it's later than I thought. At the end of the day, my brain is cooked. Right now I'm motivated by the novelty of this new tool that gives me superhuman productivity, but a few years from now when that novelty has worn off? Well, probably a few years from now we won't have software in any recognizable sense anyway, so maybe it's not worth thinking about.
At best all we can do is make knowledge work cheaper right now. But knowledge work was never the bottleneck behind energy prices, food production, transportation, cargo, etc.
It kind of was though? It was knowledge work that brought us fertilizers, tractors, trucks and large ships, which in turn made food as abundant as it is today.
Otherwise most of us would still toil on the fields just to harvest a tiny bit more food than our immediate family needs.
> It was knowledge work that brought us fertilizers, tractors, trucks and large ships, which
Yes but the number of people who actually worked on the R&D and the amount of time they spent on it is absolutely insignificant on the grand scale of things.
Oh but AI has wiped out some of the most tedious work I have to do. I can now easily work every waking minute, and some sleeping minutes, at a level approximately comparable to my prior 8h workload from 10 years ago if I wanted!
The amount of mental energy saved, and downright stimulation gained from seeing rapid feedback in a tight loop is making my workday much easier. I have enough left to continue it longer.
EDIT: Making the continuation mandatory would be a bad thing, but it's not bad that I can do this now.
I’m of two minds on this. Yes I get a lot more done. Yes I create a lot more bugs. Yes I have to spend a lot more energy understanding some things the AI does vs having written it myself.
It’s a mixed bag. I would say the worst part is that the increased level of output AND the increased level of tech debt is becoming accepted as the norm.
I find it enjoyable to have the energy to think, and less cognitive overload by firing off a suggestion to be checked overnight.
You're right that this would be awful if mandatory.
but it often was socially mandatory!
At least when those times come, the work can go much further before I have to actually "check in" vs doing some well-trod analysis via automation I've set up. That's great!
I think there's antibodies around disagreeable AI takes, honestly. It's fair enough. We're in this funny don't ask/don't tell stage still, regarding AI use.
Jeez, great place for discussion about new tech then. What precisely went wrong - me expressing enjoyment at finally being able to try new ideas, statements about my actual work hours, or diverging from consensus?
And what does this have to do with psychosis - is that a general catch-all term for using AI too much?
I always plan our capacity based on a % of bugs/support/upgrade vs. new feature development which is roughly at 50%.
AI will be able to reduce this while also increasing quality which will allow us build more features faster which is actually quite fun.
How it will impact budget etc. no clue yet, in best case though i can transfer my team into a new product while maintaining the old one with very little effort.
If the overall efficiency of the economy improves, "the same amount of money" can afford more things. You don't get a bigger share of the pie, but the pie gets bigger.
If you also want a bigger share of the larger pie you need to be one of the people driving its growth, merely being employed by one of the many competing companies that is purchasing more efficient tools from a third party vendor does not qualify.
there is a 3rd option and it involves us organizing and actually exerting some of our power as the critical dependency in basically all of the modern economy.
Um… work is not important. Value is important. Your employer is employing you because they expect your value to be higher than your cost. Moreover, if you are busy, your marginal value is likely declining. You are not taking the time to think about things that will actually have high impact. This to me is the biggest problem I see with AI assisted work, it makes it very easy to feel productive, when in reality you only worked on the easy low value things.
What’s more, is I have a very distinct experience working on a team years ago that decided to do less work. The team actively started turning down tickets and commiting to fewer tasks each week. And the counter intuitive thing, was that not only did that force only the higher value tickets get closed, we actually ended up closing more tickets. Was really wild to watch and experience.
1. Clankers do almost everything for us as ethical slaves
2. Everything costs next to nothing, from raw materials to finished products, and much of it can be done at home, sustainably and even better than we have it today
3. Society transitions towards utopia, basic needs met for all, away from hand-to-mouth, and we spend our time on friends, family, community and personal enrichment
The alternative is a dystopia beyond Orwellian.
People need to get this idea into their heads: our governments work for us and with almost infinite cheap robot labour and national resources the focus should be on society and not enriching some fucking crooks.
Its literally us or them, literally a fight for our lives
> our governments work for us and with almost infinite cheap robot labour and national resources the focus should be on society and not enriching some fucking crooks.
But also, with almost infinite cheap robot labour (and army), our governments no longer need to work for us.
And this is where we go full speed, and if anybody voices concern there is 'think about children!' for 'em or just laugh them off as luddites. If you ever wanted to move social layers, yesterday was too late.
But imagine the control - how desperate folks will be for some pocket money if half of population is unemployable. So much control! So many cheap gardeners / sex slaves to choose from! You would have to work outside openai / anthropic to refuse that, for sure you will be one of few on top of the pile, not some trash within it right.
I think there is a real world space separating people in work experience.
I've never worked a job where 40hrs was even enough, so any efficiency is instantly absorbed. That is the work world as I see it.
However I've run into in people online and in real life whom's entire work life seems to be on the opposite side. It seems to be a rather large swath of people as well. They have never worked anywhere where they even have 40hrs of work a week to be done.
Even last year I ran into a guy who gets paid 140k to by his own admission(12+yrs), to basically update a handful of websites a few times a week(shockingly...he is concerned about ai taking his job). Those jobs are not coming back and those that exist are rapidly drying up. Thats the disconnect. There are in a national sense a large number of people that have literally never worked an actual job even though they went into an office their whole working life, and everything about their life outside all reflect that they have worked a job despite never having done so. Those people are in for some tough times.
In my experience, at some places you can also have both: a job where during some weeks there is little to do, but you have to spend 40 hours sitting in the office regardless, and during some weeks there is so much to do that you need to work overtime. You need to be flexible, you see, but the flexibility only goes one way.
In other words, companies capture all the benefits of AI, while the workers get the downsides.
No wonder there's so much animosity towards AI -- it's not the tech itself, it's the fact that it makes the lives of a small minority better while making the lives of the vast majority worse.
It could be used to make most people's lives better, but it's not, and it won't be.
It's not that simple. In a time of massive flux and shake up you have to be ready to jump to different kinds of roles in different kinds of organizations. Maybe you've been an enterprise and need to jump to a startup or the reverse... or maybe you've been in an internal only role and need to move into a customer facing role or any number of other possibilities. Admittedly one way of interpreting the need for evolving is that it could be exhausting so in that sense the author has a point but it's not as black and white as you imply. At any moment in time there are growth areas in the economy and growth teams within companies and there are atrophying or declining areas of the economy or teams within companies. These macrodynamics are just as important as who you happen to work for and what the tools du jour mean for your work. But they require you to be intentionalable about making changes
>you have to be ready to jump to different kinds of roles in different kinds of organizations.
How did that work out in 1933?
There are limits to how much an economy can shake up before structural problems appear that means when the music stops there is no chair anywhere for you to sit in.
>At any moment in time there are growth areas in the economy
For bread line server?
Worse with the modern world you're not just competing with your neighbor, you're competing with the entire world and if the other parts of the world have lower costs you can quickly find yourself in a trap.
The poverty of the Great Depression was put to use to great effect building so much of the infrastructure that we now rely on. It was bumpy that's true. But it's not accurate to imply that bread line work was the only growth areas then.
Oh, but it's very simple. You have elections, you collectively don't vote for the uniparty, you raise the minimum wage and then they will pay.
One can pretend it is very very complicated and invent many excuses. One can pretend not to notice only the uniparty is promoted on the Tee Vee and in the old media. One can pretend it doesn't cost many millions to pull the wool over your eyes.
Surely if increasing the minimum wage was impossible they wouldn't need to do all that? They would keep the money in their pocket. Apparently those with absolute control over everything experience it quite differently? They seem to still believe you can change things.
>The opening premise baffles me. What kind of illusion must one be under to believe that employers would ever pay you the same amount of money to do less work?
Work is about the work written down in your work contract. If you manage to fulfill your contract by doing less, you still worked according to your contract so you should be able to expect the same amount of pay.
Let me get this straight: If someone who works for you takes more time but achieves less in that time, you would still pay them more. And if they took less time but achieved more, you would pay them less? That's some sick logic.
> The opening premise baffles me. What kind of illusion must one be under to believe that employers would ever pay you the same amount of money to do less work?
You do much less work than a farm labourer 200 years ago and get paid more.
We know what this looks like, over all of human history, and it’s not 300m people living comfortably on checks from the government.
We do need to do something, and soon, but “tax those guys a lot and send the proceeds to government services or pittance checks” isn’t going to cut it.
Wealth redistribution doesn't have to be 50% tax rate for billionaires for their current wealth you know.
If can be just factually and morally correct step of taxing given company for income from given country and thats almost it. Maybe close few biggest loopholes how to hide income while having massive revenues. None of the faangs operate like that, with such move their revenues and stock prices would fall down since part of valuation is based on continuing this tax scam/fraud.
This would spawn few hundred billions globally to start with. Of course tech bros et al would launch some massive PR counter-campaign also here, thats all fine and expected.
Will it solve everything? No but will set a better path for future while helping also right now massively. If you have kids and cared even a little what world they will inherit from us, this is how you build its foundations. Not hoarding money at all costs and fuck the rest since everybody relies on rest for existence and thriving.
Technological progress is multifaceted. Some of it is more value-neutral and tends to be a general accelerant.
In terms of social effect, there's two axes that jump out at me for things that are _not_ value-neutral:
- technology of violence, where the axis is whether it makes it easier or harder for some people to have direct power over others, and
- centralization/decentralization technologies.
I think there's a decent argument to be made that the former category is mostly a subset of the latter anyway.
In a _very_ hand-wavey, un-nuanced take, decentralized systems shift power and profit away from the establishment and are inconvenient to use, and centralized systems tend to be easier to use -- and they support and create establishments, and give them power and profit. Unsurprisingly, a fair number of establishments exist and grow precisely because they found a way to centralize a decentralized technology.
The choices that we make in our lived systems mean these axes are mere tendencies, not predictions. The "old" Internet was largely a decentralization technology. The shift towards platforms (Facebook, Xitter...) made many of the old-internet functionality easier to use by centralizing it. Blockchain is superficially and deliberately a decentralizing technology, but with the choices we've made, it's mostly empowered people positioned to centralize it (initially in the form of fiat currency exchange points).
LLM progress is currently in a centralizing phase, but overall its direction remains unclear; will open-weight models running on cheap commodity hardware (once OpenAI won't or can't clear the aisles of available VRAM) turn the tide, or will the frontier labs retain a permanent structural advantage -- and what market and regulatory choices will we make to nudge that direction?
By your logic, the quantity of work will eventually expand to the point where 1 person (or less) does everything. I agree with you.
I did a deep dive with AI last night and asked it: if we extrapolate the level of industrialization vs the rate of job loss due to AI, what year does it become cheaper to live off-grid in an independent community?
The AI's answer? Between 2030 and 2035.
Meaning that money will soon require so much productivity to acquire, that it will eventually become easier to produce goods and services independently and provide for ourselves.
The main reason is that currently labor represents 60% of the workforce while owners represent 40%. That is flipping to 45/55, meaning that labor is on the verge of losing its majority stakeholder status and collective bargaining power.
When that happens, we'll only be left with market forces propping up wages. Meaning that AI automation will gradually undermine wages to the point of subsistence.
When that happens, the opportunity cost of exiting the traditional economy drops to 0. Why would anyone work at the behest of an owner when they could be an owner themselves for the same or lesser labor?
Now, we could reject that and pretend that supply and demand curves won't be undermined by a falling cost of production. We could do that, but I'll leave that as an exercise for the reader who won't hear a word I'm saying anyway.
If we stay rooted in history and extrapolate, we're left with at least these options:
a) Allow free market capitalism to regress into authoritarian neofeudalism and create conditions similar to those of the early 20th century when robber baron industrialization caused widespread wage stagnation, leading to revolution and the seising of industry by displaced workers (communism or more accurately state capitalism like Russia's Bolshevik Revolution in 1917).
b) Raise taxes on the wealthy/capitalists (the owners of the means of production) through increased regulation, and use the proceeds to buy back and nationalize those economic inputs to fund public services and UBI (socialism like China's Second Centenary Goal or 2049 Plan).
c) Embrace the rapidly falling cost of technology to get ahead of widespread unemployment and bring about something like solarpunk, where open source tech like AI and 3D printing allow society to reduce the cost of living to nearly 0, so that people have the leisure time to embrace art, culture and the humanities to finally achieve widespread self-actualization (post-scarcity economics a.k.a. Star Trek when it eradicated poverty, hunger and war around 2113, 50 years after Zefram Cochrane achieved warp drive in 2063 and the Vulcans made first contact, as the story goes).
I'm partial to one of these, and have been putting my efforts into a home economics that can provide for at least one other person so that it can go viral. If we dream big enough, we could finally have what the wealthy and powerful will never have: enough.
> The opening premise baffles me. What kind of illusion must one be under to believe that employers would ever pay you the same amount of money to do less work?
I guess the idea that AI will do all the work and either take care of us or kill us all. Which will probably happen, but the transition takes years and in the meantime we work much harder. People were expecting a fast takeoff.
>Not because any single task is harder. Because there are so many more of them, all at once.
There's a nuance here since the author says "Almost all of this is private work," but this is basically the entire premise of Cory Doctorow's excellent The Reverse Centaur's Guide To Life After AI, which everyone should read. If I may quote:
>>In automation theory (the academic study of automation), a centaur is a person who is assisted by a machine. Think of a clever human's head, arms, and hands atop a horse's strong body. Riding a bicycle or driving a car makes you a centaur; so does using the mute button on your TV remote when an ad comes on. Wearing a hearing aid makes you a centaur, and so does using a calculator to multiply large numbers. Being a centaur can be glorious.... Best of all, I get to choose exactly which of these features I use.... No one expects me to write more pages when I get a new tool.
[snip]
>>A reverse centaur is a human who is conscripted into acting as an assistant to a machine. There's a classic I Love Lucy episode where Lucy and Ethel are working on an assembly line at a chocolate factory, taking bonbons off the belt and wrapping them in paper. As the belt goes taster and faster, Lucy and Ethel have to work at superhuman speed. They're reverse centaurs: the machine can move the chocolates from one place to another, but it needs a human to pack them into the box, and the humans who act as its assistant are made to work at a pace that exceeds all human capacity until calamity ensues.
"More tasks all at once" is exactly the premise of that I Love Lucy episode.
I'd think about it this way - with LLMs I'm still a human with one brain, one set of hands, one keyboard, one screen.
I can delegate more quickly, but still I'm constrained to what I can respond to because I can focus on limited amount of thing at a time. It's like I was given multiple sets of hands (so your centaur parable stands), but that doesn't make me faster if I need to do multiple things at the same time - have you tried juggling, chewing gum and skipping at the same time?
+1 to The Reverse Centaur, but I don't think it matches what the author describes. As a VP of Technology he presumably has some latitude in choosing what to work on.
One thing I feel like a lot of us are missing is that we no longer have to worry about looking for a missing semicolon, typo, or off by 1 error, or manually combing logs to try and understand a problem.
Sure we have to do a lot more work, but i would say it is less mundane than it was. Now i get to focus on the fun parts of software engineering. I think the burnout some of us are feeling is because as the author has said we don't know how to properly use the tool.
Those techniques will come with time through us using the tools, and coming up with the standard for how to use it for the most productivity gains. For instance how many of us, when manually coding, have to have headphones with noise cancellation, or use the pomodoro technique to get things done. I imagine soon we will have similar techniques for working properly with AI.
It depends if you were a dev who liked more the micro parts or a dev who liked more the macro parts. To compare, think about going to an artisan or a musician or a painter (or any creative like most programmer who started by being passionate about it) and telling him "well now you can ignore the fine details, there is a tool which does that for you and you just focus on the less mundane tasks."
This. This is the real phenomenon of the AI-assisted development decade.
People who became programmers because they are naturally motivated by the craft, are now "unproductive" unless they dispense with their beloved crafting activities forever. What are they going to do? Voluntary buyouts for starters.
But then who will be left? What kind of software will they provide?
I want to agree with you, and my instinct certainly is to do so. one thing is for me is, I don't see code as art, to me it is math / science, and we are scientists, using the scientific method to solve problems with questions, hypothesis, evidence, and conclusions, all in the form of documentation and code.
I love writing code, it is quite thrilling going through meticulously and spotting that thing no one else could. Writing code, however, is daunting, taxing on the mind, takes a lot of mental toll, and you must be in the proper mental space. So if i have a tool, that can take some of that weight off of my mind, so i can focus on the bigger picture, is a good thing.
The end goal is still software that people use. I think we should focus more on our new found extent of our powers. I feel at least 6x my normal velocity. How many new projects have you been able to tackle since the advent of AI? I know for me it has been numerous for sure. We're living in the future man. I write code on the train, and in my sleep nowadays. Crazy times.
I agree, great coding is art too, and not just the code itself, but very much the problem solving process. It’s sad the highest levels of cognitive ability humans can do are no longer needed (or not nearly as much)
Agreed, really enjoy solving tough problems. Like having some interesting enhancement that needs to be carefully and surgical added into a complex codebase. Or writing an algorithm that grabs a bunch of info, sorts it in multiple ways, and uses this info to generate a useful result for our API (am thinking of specific past situations that have done this).
What is not fun is the repetitive stuff, where you have to search a code base for the same change over and over again, but it's not something you can just do a search and replace. Am happy to have AI do this (although, even this, you still have to read through every change).
Apples and oranges. Hot take: fine details in an artwork are extremely important. "Fine details" in a codebase are either some pattern you can easily instruct an agent to follow, or deliver zero value.
Missing the point. Which is, some people like the details, code as craft; others like the high level, product/business impact, systems, etc, code as a means to an end.
The profession used to have a need for both, now the “coding machine” and artisanal code crafter are trending obsolete in most companies.
I think I reacted exactly to this. What is "code as craft", or as often read, "code is poetry"? It is variable naming, structure, hierarchy, import logic, all patterns that can be strictly controlled still. Opinions vary if this still carries enough value - I like these details, exactly because I value the "craft" and the expectancy of human oversight these presume, but I could argue about it to some manager less and less (compared to what frontier models generate already).
So when all this can still be kept, "hand craft", the muscle memory of typing code, or, as OP said,"dev who liked more the micro parts", what is this other than some nice myth the industry wove for itself?
This 100%. Past me would really enjoy writing a regex or inspect some undocumented API. Today me likes solving problems more, so I just throw Claude at the undocumented endpoint and after 10 minutes I've got a documentation with examples.
It's completely true that we get to focus on the fun parts, but I think the burnout some of us are feeling is that the boring parts were also protecting us. It wasn't always fun to spend an afternoon mapping out a 500 line python class; but it sure was relaxing, and a good reason for deadlines to shift forwards and back by a day or two when they needed to. Now we only do the fun parts, which also are all of the least well-specified tasks, the parts that require 100% of our brains, the people-focused parts, and the most time-sensitive parts. All things which leave us tired at the end of the day, even if they are fun. Not to mention the 5-10x increase in context switching, which is itself draining.
Perhaps we'll work out new norms for how to organize our time sustainably. Or perhaps ClaudeCodex will finish blowing through the job of a SWE in the next 24 months (remember, it's only really 12 months old in its current form).
It also speaks to a broader shift away from learning culture in the profession. Mapping out that 500 line Python class may not feel productive in the moment, but it builds understanding that can lead to deeper insights later.
It seems like agentic coding is removing the last few places where work really did force us to slow down and spend some time learning. I'm not convinced that's a good thing. I've seen too many situations where the rich mental map someone built up during learning time later saved the team multiple person-months' worth of effort by letting someone see a non-obvious solution that was much simpler than the original plan.
> One thing I feel like a lot of us are missing is that we no longer have to worry about looking for a missing semicolon, typo, or off by 1 error, or manually combing logs to try and understand a problem.
This is what I hope for, but every time I use ChapGPT and similar, I get results which take work to optimize, so it saves me little time, if any. I admit I don't use paid versions of the A.I. stuff, so perhaps this is the reason.
I have found, that the best way to prevent bad output with AI is proper planning, and also extensive end to end tests.
- If it is implementing a design, I have it compare the screenshot with the mockup and get 97% similarity.
- If it is a functionality, I have it spin up a browser/tv/mobile and check the output is what i asked for.
I also use plannotator for planning, that is an amazing tool! need to buy them some coffees for sure.
The paid versions aren't actually better than the free ones in that respect. They still make mistakes more often than a human would, and you still have to carefully review what they put out. The only way to get a speed boost from LLMs is to give up on the quality of your software and just YOLO.
> Now i get to focus on the fun parts of software engineering
I enjoyed those parts too, for multiple reasons. Typos often forced me to ask a colleague for a fresh pair of eyes (who'd almost always spot the problem immediately. Humbling, and great for team cohesion) manually running git bisect to isolate a problem helped me understand how different changesets interact, making me better at reviewing code. Debugging race conditions taught me resilience, and deepened my understanding of the code base, and the satisfaction from fixing a gnarly problem was fantastic. I also enjoyed gaining mastery and becoming the go-to person for a certain class of problem. These are all gone now, sacrificed on the altar of efficiency gains. Now it's "BTW, I noticed a subtle race condition while refactoring variable names, and fixed it. Review my 6,000 line diff now please."
The unexpected variety of activities gave texture to my working day, and was probably good for the brain. Does anyone actually achieve flow state while using agentic AI? I don't[1], and I hate it.
1. except occasionally during design, which isn't thr same. AI side quests inevitably overflow detailed mental models.
> we no longer have to worry about looking for a missing semicolon, typo, or off by 1 error,
These are all things that static analysis tools do very well. I have been using linters and type-checkers for almost my entire career that caught these things. For the past 10 years at least these tools have been integrated in my editor such that they highlight a typo, a missing semicolon, etc. as soon as I type them. This was never a real problem for me, and I assume most experienced software developers.
This post comes across as a carpenter who rejoices in the fact they no longer have to sharpen their tools because now finally, they can buy all their interiors from IKEA.
I notice I am doing more work as well but enjoying it and taking more time to recharge. I stopped listening to podcasts and audio books while working because before, my linguistic brain had time to linger while puzzling. Now I’m reading and writing letters and memos all day in addition to that other puzzle solving mode. No mental bandwidth for those extra things.
And multitasking is easier with ai. Have a lead agent keep track of your work. With a real task tool. I can drive several projects at once. I finally (have an agent) keep my work journal updated.
I do miss raw coding sometimes. But if I am honest most of what I did was filling in boilerplate code that was not really new or interesting. The real thinking was not the code itself. The code was a means.
I used to worry about who will learn coding anymore but seeing how I can push into new types of systems now, my worry is reduced. Kids will just have really awesome assistants who can explain it all in the way they prefer.
> Have a lead agent keep track of your work. With a real task tool. I can drive several projects at once. I finally (have an agent) keep my work journal updated.
Do you have an agent you can share for this? Sounds like something I'd like to try
I don’t really use off the shelf agents. Usually just string together what I need. In this case I give the lead agent an mcp tool and tell it to use the service to track all work, and what should count and not. For work I use private GitHub project board and for personal stuff obsidian and markdown.
I don’t like big fat agents that try to do too much. Better to keep it lean, each agent prompt is really just a short brief, like a paragraph or two. If it’s longer consider breaking off skills and tell the agent the list of skills it might use.
Notice the linguistic thing while trying to dictate prompts. I can apparently type and listen to podcasts easily but not talk with podcasts on headphones. Brain just nopes
This precisely - AI takes away your leverage as a software engineer. AI models are very well trained at coding today so I'm not sure what further improvement really looks like. But even if humans stick around to guide the AI, I expect the job itself will get more unpleasant because capital/management have no reason to treat this labor class well anymore.
Honestly I think that is more of a symptom of how slapped together the AI software stacks feel to an outsider.
All my dev friends swear this is the future, all my scientist coworkers just say it's a useful tool.
I have heard the pitch and have people explained how it works... But I have yet been handed a guide on how to set myself up properly to get the most out of AI. Like I have both a corporate openai gui agent as well as cli setup with a monthly allotment of tokens. But there is no clear standard guide on how I should be setting things up, like we do for python in books "here are the building blocks for python"/aka learn python the hard way
:D As a person who can do both ( and keeps logs of various interesting mishaps in our little corner of the universe ), neither is easy to do well, both require skill and, well, current literacy rates don't exactly make me optimistic.
edit: It is actually kinda funny. One would think that if we would embrace the full spectrum now, we could have the best of both worlds ( logic and language side by side solving humanity's problems ).
Just remember, it's not how good you are at your job, it's how hungry the person on your way to work is that's willing to split open your head and steal your wallet.
Trust me, the concept of social insurance is not lost on me. Frankly, the fact that we are running into this without considering likely outcomes ( or worse, considering it and plunging without relevant ways to keep people from going crazy ) is.. annoying to me. It shows me our leadership is.. not great.
AI also gives employees much more leverage than before, as they no longer need a large team to compete with BigCos at knowledge work. It's never been cheaper to start a software business.
The side effect of this is also that the value of software itself has never been lower, so starting a new software business isn't very attractive in today's market.
the cost of developing software has never been lower. That is maybe 10-15% of the value in a software business believe it or not. If you can do the other 85% it's a great time to start a software business.
It cheaper to create one-off products. Creating a sustainable business though? Charging money for the product or service? Actually getting enough income to thrive? I think that also just got a lot harder. Sure, you quit and you undermined BigCo with your open source release, but somebody with even less to lose will now undermine you.
Tech is the least important thing even for a tech company. Unless you are a 10x better than anyone else in the game. This did happen once for a while but you can do it without AI anyway — and probably better for you because it’s a lot more difficult to copy yours. However, hardware oriented companies may be the outliers.
The ability to rent cheap AI capacities, the ability to push to a million consumers, the ability to persuade for large contracts, these are the most important ones.
I wonder how much companies would pay people to NOT replicate their products. Like, if I'm just hacking around with my LLM, and am approaching a near 100% drop-in replacement for the entire Atlassian suite--how much would they pay me to NOT do that? Is that worth 1 livable wage to them?
Well... I think the converse to this is the companies that WOULD pay you (slightly less than they currently pay Atlassian) TO replicate those products.
I expect some BRUTAL competition for these larger software companies. My company just recently replaced our CRM with an in-house tool. (Of course, the maintenance / buildout of that is costly, but it's eventually going to be less than we are paying for the third-party CRM)
Honestly, I cannot personally decide whether or not this is Jevon's Paradox in action or not. In theory, that software could be entirely hands off (Claude write me a 1:1 implementation of Salesforce CRM and make it never go down) but in reality, someone is going to want it tweaked, modified, changed, and thus we now have an internal CRM that we own that requires some level of expertise. Maybe it's not a software engineer, but a software engineer is probably the most well equipped to take that process over...
The coding is hard, unless by coding you mean slapping together CRUD forms. Many businesses are undone or fail to grow properly by neglecting this. This refrain that everything about starting a software business is hard except the coding needs to die.
If the coding was nothing, every good marketing guy could go solo and start large software businesses. It does not happen. You mostly see successful technical founders, or usually a marriage of technical and business cofounders.
Finding a fit between viable business opportunities and the tech to address it is what's hard, and requires several skills: technical, marketing, communication, product design, etc.
The loss is from model training. Serving an open-weight model is profitable and these models are already good enough for many tasks. If Anthorpic/OpenAI will try to recover training investments with higher prices they'll loose the volume (and given infra spending commencements they need the volume). So I don't expect any significant price increase.
From a higher level view, and not just about the article I note that:
- Automation seems to make specific individuals work more, the leverage makes the individual more valuable in the short run (eg if I can produce 10x the engineering, for a little while I'm 10x more valuable, I dont mean LoC but engineering value). But in the long run its deflationary on the value of what I do, I might have to work harder just to stay in place.
- Automation/industrialization spins the hedonic treadmill. Economists ~100 years ago thought we'd all work like 5 hours a week or something because, for example, 95% of the farmers would be replaced with machines etc. However what they didn't/couldnt account for was us increasing our consumption as fast as the productivity improvements. Before today I had use for approximately 1 GPU (gaming), now I'm probably using dozens or hundreds. George Monboit also notes this with "green" technologies, that we end up consuming more such that we approximately end up with the same pollution (co2, plastic, whatever).
- Software has long been a negotiation between Manager/Executive class and the maker class about what an appropriate amount of software per week/40 hours was. They think because claude writes code 95% faster we should ship 20x the software. But they miss that coding was only 10-25% of the job. (numbers made up to illustrate idea, someone chime in with the data please!)
For decades tech companies have promised that devices save time....but to do what? The pitch was that they would save time for us to do more 'creative stuff'...but I think that pitch and idea is fundamentally flawed. From doing creative stuff for a couple decades, I can make a few observations:
(1) More tech doesn't save time if it's developed in the direction/dimension of more information flow. Because it seems that the best environment for creativity is a somewhat restricted environment of information. And if one needs to engage with information to take advantage of this new tech, it actually impedes creativity.
(2) A lot of the grunt work that we seem to be automating also encourages creativity. From what I've observed from many artists and programmers (I hang out with both), it seems that creativity strikes best when it's not forced, and the "downtime" doing grunt work actually accelerates it.
So I think that AI breaks these two principles (minimal information and the removal of the crucial grunt work), and actually makes us work more for less.
The only gains seem to be those short-term kinds (a program can be made faster, rewards can be reaped by senior people who have thought about stuff for a while), rather than encouraging a healthy environment of creativity....
The key is the information environment and its reshaping, as you noted.
Every technology, even technologies that are, from the limited and naive perspective of the user, "free" (LLms can do what I used to do a lot faster than I can, and I don't need to maintain them right?) is not actually free. All technologies bring about their own demands and need for maintenance. Often they don't actually eliminate labor, they just shift it to new areas.
Consider LLMs. Even at the user level they have created their own information ecosystem and new grunt work and overhead. Now you have to prompt, you need to organize and manage context, skills, memory. You need to keep up with model releases. You need to be more vigilant reviewing things, the list goes on.
At a higher level they create even more toil, we need to maintain vast electrical supports and data infrastructure, data needs to be amassed continually for training, the information environment needs to be filtered and sorted more than ever as it gets flooded with slop.
This is the grand irony of technology. It's actually pretty rare for it to just be a total labor saving device. More often its real benefit is rearrangement of the world's structure and resources to move labor to more optimal places for certain needs (for example water supply systems).
This rearrangement also often rearranged financial resources and power. The big problem with LLMs today is that this aspect of the rearrangement is one of great concentration, and it's concerning (employers having a lot more power to extract labor from employees is one example)
I'm more productive, but yeah, not working less. It's really going to come to a head when our team is due for a raise and we've done 4-5x* the work with the same team over the past year.
* That number isn't vetted, but it's my guestimate.
AI is increasing productivity. Companies are positioned to leverage this, employees are not, as they'll still work the same hours, maybe now producing more.
The way to better leverage this is to build your own company, I know this is not as simple as saying it but the gap is widening and it doesn't look good for employees.
As someone with their own company, AI doesn't magically make things better. If anything, it makes it worse since you have far more competition from other people who are not good at programming but are good at sales/marketing.
It's like with the music industry, it got easier for anyone to make music/release music in theory. Meanwhile for the last 100 years there's mostly been more concentration/consolidation.
Yes, but you are in a better position to leverage this technology than an employee, the potential value you can extract from AI is higher. Regarding competition, software was never the moat in most cases, so I don't think more software equals more competition, at least not linearly.
I may not work less, but certainly different and better.
Just this morning I went for a long walk on the beach, barefoot in the water. Using nothing but voice interaction, I handled a couple of outstanding CRM tasks, sorted out some accounting issues and upgraded my search infrastructure (SearXNG) and most of all enjoyed my walk outside. I would not be able to do all this without AI and would be stuck behind my desk, daydreaming about going for that walk instead.
With the power of AI, soon all of us can be doing outstanding CRM tasks instead of wasting time enjoying nature. Imagine the time savings we could recover by removing all quiet contemplative moments!
I’m a bit puzzled by this kind of comment, which pops up every time someone mentions getting some work done while out in nature.
What it seems to miss is that we get to walk around while doing some work, not the other way around. The alternative would be being stuck at a desk or in a meeting room, not quietly enjoying nature on a day off.
What is the reason you couldnt use your laptop on the beach to do work in 2019? Probably mainly that you are either working or enjoying nature, not both, and if you're working, then an office with reliable Internet, no wind or sun in your face, etc, is more suitable/efficient.
Put another way: if my twin and I are both working at competing firms doing the same job, but I do it via satellite phone while hiking the PCT and he does it from in front of a keyboard, there is no way he's not getting more done, even with AI. The bandwidth of voice/text to speech is a tiny fraction of visual.
The counterfactual is impossible to prove here, but I think it’s at least plausible that this poster, who I guess lives near the beach, could have enjoyed their morning coffee on the beach
Yea, it seems that the OP neither did better work, or actually enjoyed the walk (I mean, how do you enjoy walk if you don't allow your thoughts to wander?)..
It may be more pleasant to work outside -or it may not (it's almost winter here in Alaska -that would be a hard pass for me!). but no -it's definitely not the same, as you pointed out.
For sure, and a good part can already be automated. But certain things still require human intervention and always will.
Keep in mind these work related walks do not replace other outdoor activities.
I get the idea and I used to try and optimize my computer time this way - working in the park, working while walking, etc.
In the end it was just a hassle and a distraction. I’d rather have 5 solid hours of work and 5 hours of non-work, rather than 10 hours of an awkward hybrid.
The type of work indeed matters a lot, if you need any visual confirmation or have to look things up yourself it's probably not worth it.
As a solo entrepreneur I manage my entire business workflow. Everything is accessible from an API, from accounting to my CRM to S3 storage or my PostgreSQL cluster. Tasks can be handed off between different agents to keep things separated.
Not sure when you last tried, but I feel I would not be able to do what I do today with last years models.
For many of us, this would not be a better experience. The physical and mental components of doing it yourself is half the pleasure
Of someone’s carried my ass around all day I’d be a physically atrophied and mentally addled maniac, on obese to boot if I’m stuffing my face the whole ride
Part of the dilemma with modern culture is that most people seem to be striving for and hoping technology will provide pleasure funk laziness and glut. But these things in and of themselves produce bad outcomes. I get why people really stressed out might daydream about that, but this is probably more indicative of being overworked than of the virtues of laziness and glut
For those of us stuck in an office building, at a desk, AI models have enabled handling multiple tasks as a same time as fast as we can think. I'm absolutely doing multiple times the work I was doing before and the time between tasks is basically zero. Same pay, though, and drastically increased productivity and sense of effort.
Or you would be unemployed and nervous about how you were going to pay your bills, replaced by someone willing to do the same work for lower cost while chained to said desk.
What seems better to you, seems a bit nightmarish to me. If I'm going for a walk, the last thing I want to have on my mind is dealing with CRM software in any way.
But it's a purpose built, tailored to my needs CRM, mostly automated. Like I ask, "which outstanding issues need my attention". I tell it what to do next, when to notify again and I'm done. All tedious form entry work is taken care of in the background.
The choice is do you want to walk or sit behind a desk when doing a task that needs human attention.
It doesn't have to replace a nice "offline" outdoor activity.
But isn't the next step where your employer tasks you to codify all these tasks and capture your "process" and "reasoning" and then you can still walk on the beach all you want but you won't have work to do. Isn't the next step for a junior engineer or a single verifier to just add a minor human verification step at a higher level to remove you from the loop? I feel like there is not enough discussion about this mindset shift. Early adoption is taking away the employee moat that some people have but we don't seem to be considering that it is going to be recursive and repeating.
It reminds me of the 7-minute abs scene in Something About Mary.
> But isn't the next step where your employer tasks you to codify all these tasks and capture your "process" and "reasoning" and then you can still walk on the beach all you want but you won't have work to do
No, the next step is the employer to ask "why are you walking on a beach when you could be working?" and won't accept the answer that they are just orchestrating agents.
The only part the employer will focus on is "walking on a beach" and think "I don't pay you to walk on a beach."
(Caveat: OP might be running their own business. That's different. Also makes me rethink my life choices a bit).
Vertically integrated solo business on a Patroni controlled PostgreSQL cluster spread out over multiple dedicated servers with multiple providers.
Shocks Opencomm 2 headset or my Sony xm6 earbuds connected to a custom Android app. Android local Supertonic model for text to voice, private whisper on the server for voice to text. Local Nvidia Glider LLM for PII filtering of CRM related data. The Android app controls codex app server instances (mostly working with latest GPT model on high), switches between projects and threads based on voice commands, a ton of little tweaks to optimize my workflow. Every service except the public facing website runs behind a Wireguard VPN.
Yes of course your employer is fine with you going walking barefoot on a beach instead of being in the office working, and it's happy to pay you the same money. Come on...
Also, you are telling me that you can get only with voice interaction an AI to write well engineered and maintainable code, something I've yet to see an AI be able to do, to me the result that they tend is at best the same of a junior developer, except that the junior developer learns, the AI continues to make the same mistakes (because it's context is limited, when it's full it's full).
Honestly, skimming through it, the writing style made it pretty obvious to that it had been written by an AI. Pangram also indicates that the text is "100% AI-generated." The disclaimer isn't even honest, I doubt AI helped to just build a "draft" of this article...
I know, this ship has long sailed, but I just can't get used and accept this kind of writing.
AI promised less drudge work, and this is a natural outcome of that. For all that we complained about drudge work, it had a redeeming quality that it was more cognitively lightweight and gave our brains some latent downtime.
It may not seem like it in the moment, but writing code to specifications, or filling out reports in a given template, or formatting power points, was straightforward enough that it gave our brains some respite between cognitively intense bouts of synthesizing what we were about to write down: diagnosis, analysis, problem solving, "stakeholder alignment", design, architecture.
The act of generating the actual durable work product (code, reports, documents, spreadsheets, slides, etc) is largely just translating the results of this cognitively intense phase. It may not seem like drudge work, but it is a lot more mechanistic and hence cognitively lighter, even meditative in a sense. The separation of phases is never that clean, of course, but I think you see what I mean: writing a for-loop is muscle memory, figuring out that you need that for-loop is the hard part.
Now with all that drudge work being done by AI, all we're left with is the hard part. Which is why it feels more stressful: because it is.
AI companies did promise that, and for some time back in 2024 and early 2025 it did feel so to me. I just had a kid and suddenly could not work the extra hour every day I used to, but kept producing as usual.
Nowadays I’m receiving longer emails than usual and longer code to review. Less spelling errors, but long and pointless prose. Less superficial bugs but more sneaky, fundamental problems.
There is also the trap where you go from 0-80% immediately but end up spending time anyway fixing the last 20%, and it’s unclear if there was a benefit or loss in productivity. Sometimes the almost there rabbit hole of prompting can be misleading.
I've done it a handful of times, I'll get a little ambitious... "Oh, 6.1 Sol can totally handle this" and have it spin up a project, keep going, keep going.
Eventually, I get to a point where I start to do the homework and I have to step way back to "fully understand" the work from the ground up... It sucks, but I've largely stopped losing time to that.
For me, "pair programming" and incremental changes beats the "one shot" approach every single time.
We cannot seem to skip the stage of understanding (as much as some people on my team want to!)
I recently made a change because I felt the same way. I think it was primarily my phone. I had my full agent setup accessible via my phone and was able to be as productive as being on my machine wherever I was. This made it so any thought, or tickle of the mind resulted in me whipping out my phone, firing off a prompt, and iterating until it got it right.
I switched to a dumb flip phone as my primary device and now my work mostly only gets done on my computer during work hours. I obviously still have my iPhone and I sometimes take it on walks during the day if I want to have a productive walk, but it mostly stays plugged in in my drawer.
I always thought I wanted to be constantly connected to my dev environment and be able to work from anywhere, but it turns out that adds way too much noise to my life and is a net negative. I was constantly thinking about work and then able to act on those thoughts so the cycle just continued until I fell asleep.
> I obviously still have my iPhone and I sometimes take it on walks during the day if I want to have a productive walk, but it mostly stays plugged in in my drawer.
You and I have different definitions of a productive walk. To me a walk is productive if I get my heart beating fast. Bonus points if I see an unusual bird or insect.
I’m getting to that point of switching to a dumb phone. One thing I’ve done this week is completely not allow myself to respond to people/do any work even 1 minute after my scheduled end time. It’s wayyy too easy to do that these days.
Hm thats great insight. My phone contributes a loit to my phychosis and bursts. I check what my sessions are doing, i spin up new ones etc. If i distance my phone at least i ll stick more to the trad 9-5 schedule.
This reminds me that I absolutely love having a voice phone conversations in other words the old way. No video no battery drain simple high bandwidth voice to voice
Technological innovations have never meant less work for the worker. It's the bosses/company that benefit from being able to squeeze you that little bit more.
Want to do less work? Join a union and campaign for a 4-day work week or something. If the productivity benefit from AI is real, there's no reason you shouldn't be paid the same for less time because your output is the same/better.
> If the productivity benefit from AI is real, there's no reason you shouldn't be paid the same for less time because your output is the same/better.
That reasoning makes sense in a world where people are paid in proportion to the value they create. But it doesn't make sense in a world where supply and demand determine the price paid for something. Because if you can do the same work in less time in means you've actually increased the supply of producing things and thus decreased its worth.
There's no real reason why we work a standard 40 hours. We just wanted it, so we setup our society to work that way.
I want to work 30 hours a week. Do you want to work 30 hours a week?
It's okay to just vote for the things you want.
So many people make their voting decisions based on, like, whether or not billionaires will leave the state, but--again--it's okay to just vote for things that benefit you personally. Who told you to worry about billionaires, and why did you let them influence you that way? The billionaires will be fine.
> there's no reason you shouldn't be paid the same for less time because your output is the same/better.
The reason is economic growth for the company / the country.
This in turn is a more subtle point than it seems at first glance. Not having economic growth isn't really a desirable option for a company, a country, or in (the long run) the people involved.
Outside of that small 20ish year window, workers have never benefitted from increased productivity compared to the capitalists that have stolen $50 trillion from the masses:
That's a separate probem that probably wont be solved by reducing the work week.
That said, ppp adjusted median houshold income has grown fairly steadily even in the US, even with growing inequality... so workers too have benefitted, just not as much as they could of the gains had been spread more evenly
My two sons are programmers, like me. We also went through a similar process after Claude Code and Codex became good enough, sometime around November 2025.
The key is, in my view: "I have several agents running in parallel [...] None of them needs me all the time. Every one of them needs me sometimes." This is what kills flow. Constant context switching plus having to understand the detail is really hard, and leads to exhaustion and burnout.
The solution for us has been to scale up. These days we can each manage on the order of 100 agents, working in different projects. I only talk to five or so coordinators, and to the ones that are implementing particularly tricky features.
Which means that I get to think, and pause, and focus again. It is not the same state of mind I had when actually crafting code, but it is fun again.
the amount of work I get done increased exponentially, but so did the number of things I have to keep track of in my head. it's tempting to just YOLO everything to the AI, but "the AI did it" isn't really an excuse when production goes down.
Definitely working as much as ever but I must say that AI does make it about zero stress for me because delivering isn't really a question at this point.
This tool will reduce your work has always been a lie in the corporate world, at best it's shifted your time from one pool to another. The only time it really works is if you secretly bring in the tool and don't tell your bosses you're getting things done in less time. When your bosses know you just get rewarded with more work to do.
Kind of surprised not to find any instance of the string "jevon" in either the article or the comments here already. Even with the assumption that AI decreases the amount of work needed for writing software (which people may disagree on of course), Jevon's paradox seems like it would still apply; maybe making more stuff feasible more easily just means management has higher expectations https://en.wikipedia.org/wiki/Jevons_paradox
I use ai as I program every day and I find it useful. I also use it for tech support for myself and my home server. The other day I used it to make some changes in my server and I felt like I accomplished tons. But when I thought about it in the evening I realized I could have got it done with less time if I had just read some documents - the ai advice was often out-of-date and I notice that while programming too - for example, I use the spring-ai framework which has changed massively between versions and the ai always gives old advice....
So on average I would say I'm getting about as much work done now as before ai.
I've been throwing it at a homelab as well as I don't have to worry about production stuff and can revert snapshots fine.
Until the agents decide to start messing with snapshots.
It can help if you can't remember where something is, but then you'll never learn so I try to action everything myself, but again agents like to just blab on constantly and overstep.
I found I was doing bulk changes etc with it, seemed helpful until I realised I could just use dsh as Ive done in the past and basically do the same thing with way less resources and more consistency.
I like to compare AI to the advent to spreadsheet software. When we look how accountant were working before computing and then with calculators and later how computer and spreadsheets changed the game - we still have have accountant but the one we have today don't do what they did 80 years ago - name is still here - function roughly the same - but the daily work is different
I really don't understand some of the comments here. I've honestly been automating so much of my work that I spend most of my days working on personal projects and checking in on what's being done on my behalf by AI. This, of course, after spending a few weeks setting up validation tools so I can trust the output.
But someone in the same business field as yours, who, instead of spending time on personal projects, is working on the business itself, will be ahead of you.
Maybe. I've been able to spin off crash fixes, CI/CD improvements, and even demoing my personal projects as AI exploration. So my visibility is at an all time high.
Same. Once I actually tweaked my agent to do all the same work I used to do by hand, I now do so little work that I often try to think of ways to performatively imply that I’m working hard out of fear that maybe someone will catch on and I’ll just get fired out and replaced with my agent once and for all.
A neat trick I’ve found is that I can just invent random work for myself in the form of building custom tools and skills for various workflows and write it off as research.
The way I tell this to people is "I have never worked less and I have never been more productive".
All of my effort goes into ensuring that agents work autonomously for days to weeks without me having to coddle them. This takes a lot of effort to ensure that my memories are pruned, my git logs are clean, that I've transferred my way of thinking to agents so that they can make autonomous decisions which are 60-80% aligned with what I would want.
And the reason I go with 60-80% is that's the level of alignment I expect to get from my product & engineering teams, in a given sprint/period. Like with them, I've accepted that I'm not the perfect solution and it's going to go through a few iterations before I shape it into what I want it to be.
If I had to constantly chat with my agents, I think I'd be burnt out too. because I find that it's an absolute drag talking to the machine.
edit: changed the last sentence, which was unintentionally judgey.
Counter perspective: I like talking to the machine. Often times I learn.
I don’t have a complex system of agents buzzing around doing my work for me. I’m sure I’d be more productive (from company standpoint) if I did. Does this not result in some inherent level of mental/skill degradation though? Or at least skill transfer away from code and fundamentals/architecture/design and towards configuring autonomous systems?
> Counter perspective: I like talking to the machine. Often times I learn.
I don't mind talking to machines. I do mind micromanaging them, and I ming micromanaging people as well, and I don't like being micromanaged.
I added that last bit because in the rare circumstance that I am constantly going back and forth with AI, I feel that I'm the one who's being micromanaged.
> Does this not result in some inherent level of mental/skill degradation though
Not in my experience. I get to try a lot more things without having to actively sit with it. In the past, I would assign it to a engineering team if it was practical for the company. If it was side projects, if it was really, really important, maybe I'd squeeze in some time on the weekends, but mostly I didn't have time for it.
Now, experiments really come down to do I have the personal or professional budget to do it. I do use a lot of the subscription still, so it's do my subscriptions have capacity. I try every harebrained idea that comes to mind.
I definitely feel I've developed more skills in the last 6 months, than I did in the 2-3 years before it.
Fearless prediction: At some point, this dynamic—-i.e., running to keep up with the AI and providing directions when you catch it or when asked by it, with little creative work in between--will slow. Without human creativity, focus, and the resulting code, ideas, text, and stories to steal, LLMs are essentially helpless.
If you're an employee, that's true - but as a founder of a small SaaS (https://arcentry.com) I genuinely work about 1-2 hours less and get more time walking with my dog through nature. I now have the luxury of concentrating on building the core product and writing articles for the blog and AI handles the things I'm less good at or passionate about, including marketing and content distribution, accounting, building the website (which admittedly looks a bit AI-y, but still much better than my handmade version), answering simple customer requests and escalating complicated ones to me and so on.
Meanwhile, I get to do more of what I enjoy. Of course, it's always tempting to let Codex work on the core product itself - but I resisted that urge so far as I feel a strong sense of ownership and craftsmanship - plus its the thing I actually like doing.
Love this, and it's the outcome I was hoping for. Did you have to deliberately cap your hours, or did it just happen? For me the cheap extra capacity always gets filled with another idea.
Isn’t it possible we perceive we are doing more work just because we are producing more output, closing more tickets, and releasing more features? Of you’re still sitting in the desk giving the same brain hours to the work, then you’re doing the same amount of work, in some reasonable sense
This is also where is makes sense for “the company to expect MoRe WoRk FoR tHe SaMe PaY?!?”
But, this model does (likely) result a reduced need for headcount. That’s the scary part. People complaining about putting in the same hours and being chaffed that they get the same pay for more productivity is bupkis. But people getting cannned or unable to find work, I really feel for them.
I work remotely and i feel my hours did go up, not just my output, because the extra capacity keeps getting filled with new ideas that i can also control from my phone and keep being kinda engaged. Its not that i am required to be physically all the time on my pc more. Its that i get to be more hours engaged? does that make sense?
I agree the headcount side is the scarier part. Especially if/when we run out of ideas or experiments.
Makes sense, but are you taking that work on independently? Like you would otherwise be just talking a walk, but not, because you CAN, you talk AI through work tasks while on the walk? Or you’d get fired otherwise?
I never thought about it like that but I totally agree we are multitasking and working multiple projects more than ever before.
At work, we started piloting Copilot inside VSCode in December 2023 and we always were slow to allow new models at work before sign off from security so in the first two years developing wasn't that different because I was still coding more by hand than with the Claude models at that time because they made so many mistakes involving large enterprise scale monolithic projects.
Now its like a switch was flipped in 2026 with the new models actually making less mistakes and able to produce more working code. This year, projects exploded because it was then possible to write custom full stack applications to replace applications like Nexus Repository and Postman that normally cost us money in licensing and subscriptions. So now I'm managing existing projects AND side projects at work to help us manage everything and save money. Then this has caused me to start creating side projects at home. This year with the Claude models the way they are has been total project chaos.
Correct. Too many people are ignorant of the mid 1800s in the US. As industrial technology improved the robber barons reaped massive wealth while the average worker worked 14 hours a day. It took literal wars to get worker rights.
We are encouraged to, and I do, use our tokens not for "throughput" but mostly for increasing quality.
We are basically not limited by the time it takes to write code, we never were. But we ARE limited in the time it takes to wrote ancillary code: CI and testing code being the biggest, but also tools.
Iv'e started on 2 bigger sub-projects that could help with coverage and finding really hard bugs that I would _never_ have even touched before, becuase I could never have fitted it with my regular work that actually has high priority.
I'm running a small software company together with a partner. We've moved from partly working with externals to 100% working with onpremise AI.
Both of us feel the AI fatigue from time to time.
My personal analysis is that the speed went up, with this the bandwith of different topics.
Other factors that weigh in are a not lowering learning curve inside the heavily iterating AI sector and the chattiness of the Agents. If not stopped they produce so much facts that it kills me. I started constantly working on this to stop the waterfall of facts and get ontop of it.
Stupid question maybe but why is a VP of Technology doing technical work? Aren't they supposed to be in meetings with stakeholders, management and their direct reports all the time? As manager your job is to delegate and manage people, not do the work (or direct an AI to do the work) yourself - unless this is a case of layers underneath the author getting fired and made redundant so more work ends up in their lap?
No cuts at all, the team's all still there. I've always stayed close to the code. Partly that's because Bold.org is small enough that the line between leading and building is thin, and partly because it's how I keep myself up to date about how we can reduce complexity and make things better in the product in general. A lot of my technical work is a lot of tickets which in the past would be in the drawer. Like small bugs, architecture improvements, code debt etc. The team focused on shipping the features and new things etc.
This tracks with my experience. The other bit that contributes here is that upstream inputs using AI also get faster - so you’re basically speeding up the entire machine but you’re still a human.
Unfortunately this means that you always feel the need to keep building or fall behind (picturing Lucy and Ethel at the chocolate factory) which means you’re pretty much always feeling burnt out and it feels harder to step away than ever before without feeling a certain way about it.
I hate it. It’s the best time ever to build. That excitement lasted about a year - but now it’s a treadmill I can’t escape and companies are not really paying more for the productivity (even factoring in the cost of my AI usage)
They aren’t paying more for productivity because they themselves aren’t earning more for productivity. The elephant in the room is that customers don’t have any more money to spend. Certainly not in this economy.
In manufacturing, an oversupply of finished product quickly leads to overfull warehouses, and the company knows to slow down production until product moves.
In software it’s less visible. Features that nobody is buying don’t physically pile up in the shipping department. Decades of accounting for all programmer time as productivity has left managers with limited ability to recognize time spent building unnecessary bloat as a loss. Maintenance burden isn’t fully recognized as a loss, either. We tend to assume it’s roughly equivalent to profit, the same as all other code the programmers churn out.
And what you are seeing now is that it was always a myth. Not everything you do at work is profitable or productive, and AI is causing that fraction to decrease. So of course they aren’t going to pay you more for it.
While that is true, I find that in a small company / startup environment you are almost always working on something that will contribute to revenue. The list is long. And you don’t have that much time for nonsense if you’re doing it right.
The other factor here is scaling of humans. Without these tools we would need a lot more humans. We don’t need as many but the ones we do have are heavily taxed. That salary savings should be passed on (again, relative productivity to AI spend) but we’ll see!
I once would have said that about startups, but lately I’ve been working at a very small one and I’ve seen with my own eyes how agentic coding rapidly undermined the company’s ability to effectively prioritize. It has basically killed the correlation between initial planning/development cost and ongoing support/maintenance cost. And it had developers implementing things so quickly that the product manager couldn’t keep up, which meant that he was basically just approving everything, including the half baked ideas, just to avoid coming under pressure for stalling the development pipeline.
I’m sure the ship will eventually right itself, but it’s going to take some time for the profession to digest how AI coding has changed things and how to adjust incumbent working norms to account for it.
"The pitch was that AI would give us our time back." <= this was never the case. The pitch is that AI will either augment people to make them much much more productive or replace them altogether.
I have a coworker who is possibly 1000 times more productive. He went from working on 1 or 2 side projects on his own pre-AI to working on 10-15 side projects at the same time, each incredibly complex and finishing them in a week. He build Android apps for controlling drones, then to 3D-rendering, then to making games, to creating various types of compilers, to creating entire AI run-times for phones. This is on top of running out of work to do so he's been doing side projects at work. He rewrote an entire team's product in a week and made it faster with less lines of code. Then he created his own language, made an AI music generator, an AI thing that compresses KV cache, multiple MCP servers for products, and then a global MCP-type server that incorporates a while slew of internal services kind of like a one-stop shop, and it has gained traction within our company.
It's insane and almost comical the level of productivity he has achieved with AI.
No what you're experiencing is productivity increase not work increase. You're doing less but producing more.
My productivity has gone off the charts using AI, but the actual effort and work required has stayed roughly the same.
A working day is still the same length and everyone still has 24 hours in every day...but what they can now do with that time has increased several fold.
Why? I’m self employed and I just work in the morning now. Can get so much done by lunch that’s there’s no need to keep going and then I read, cycle, play music, learn…
Tech never begets less work. Tech only generates more work because the mindset is always to optimize and maximize. Now that something can work 24/7, you must be available 24/7 to keep it working.
The past century of promised reductions in work fall under the assumption that competitors will mutually agree to not use the excess time to compete against each other.
No one with a brain thought that we would be "working less". We are working more efficient, so we are expected to produce more output in the same amount of time.
If this even needs to be said, OP’s experience as a Principal Engineer and VP of Technology is not a typical experience. I would expect a VP of Technology to be a very high-stress job.
But I would also expect a VP of Technology to stop doing technical work and start being a strategic leader. This is a huge mistake on OP’s part, and one that first-time strategic leaders who come from an engineering background make frequently.
You are a VP now, it's time to delegate your work. You are most valuable for your guidance on the overarching direction of your platform, not for doing IC work.
AI promised less work for given output. Same as any technological change ever. And same as any previous advancement, it immediately increased output of existing/known things.
Who is doing the promising and why did you listen to them? Did they write it in blood? These statements are strange. It's a way to place vague blame, and remove the onus of critical thought from oneself.
If your employer lets you chill, their competitors will not, and win. That's how it goes.
Prompting and reading detailed AI output all day long, trying to tease apart the goods and bads - it's extremely exhausting. You don't get the mental breaks of pressing the compile or test button, or doing small refactors like pulling a logic into its own function. Coding is gone as a pleasure at work.
Computers were supposed to reduce paper flow in business. For quite awhile the amount of printed paper was increased. This will follow that path and it will take some time if ever to stabilize.
this is so true btw… everytime i feel like oh i’ve done all my task for the week, i still end up spending most of my weekdays getting approvals and adding more.
i looking to solve this with a personal repo that will spin up claude in schedule and work autonomously while i only watch.
Capitalism doesn't work like that. The promise of "less work" is only possible if you're the only one that can unlock those productivity gains, but if your competitors are also able to access those gains, they will eat your marketshare if you just work less.
The whole post is just a deep misunderstanding of what it means to live in capitalist societies.
> Write less boilerplate, delegate the boring parts, go home earlier.
This is fundamentally misunderstanding how capitalism works.
Whoever owns your company pays you as little as possible but as much as necessary, as long as it is profitable. If they don't pay you enough, you leave. If you don't produce enough, they fire you.
There is no "they have enough now, so you get paid more or you can work less".
It is fundamentally adversarial: the whole game is for you to convince them that you are worth as much as possible, and for them to pay you as little as possible. They will lie to you for that, you should be prepared to do the same.
As for the github "contributions". It wasn't ever a good metric (because... metrics), but it has never meant less. With AI, I can generate 10x more commits, and it doesn't mean I produce more value.
AI is concentrating work. Fewer and fewer people are working like never before ("I've never worked more"), while a massive number of people are not working at all. Exhibit 1: Microsoft's layoff of 25% of its workforce this week.
The capitalist system has no intrinsic solution to this problem---in fact, it's blind to the problem. There is no problem from the capitalist perspective, this is progress! The solution has to come from outside, through public policy.
The promise of "less work" won't happen until society reorganizes around the recognition that most people no longer need to work. Realizing that reorganization will be the defining struggle of our generation.
I don't know how people are handling constant context switching - ok, I can trigger 5 agents that work on different things. One comes back to me citing a decision I've made 30m ago; then another, citing a finding to something it mentioned 45m ago; then another, then another. In every one of these cases agents are unhelpful as all hell - they overcommunicate, lie, scope creep, expect you to pay attention to every single thing they mention in passing.
You've created a plan? Fuck you - agent has exceeded the plan with 2 items. You ask about these two items? It's actually 8. Ok, what are these 8 items. Sorry, I was wrong, it was 10. And in the background there are 4 agents now waiting for your decision, where you have to familiarise yourself what they were doing and what were the decisions that you've made.
> A note on how this article was written: AI helped me pull and analyze my GitHub contribution data, build the chart and draft this article. The experience, decisions and opinions are mine.
Carefully omitted: who or what actually produced the text.
This is just a dishonest way of saying he had a ghostwriter.
> Should tech workers supercharged by AI be able to take more time off? Meta's chief technology officer, Andrew Bosworth, said he doesn't think so. [1]
> A Meta employee asked Bosworth if AI productivity gains could be channeled into more vacation time — specifically, the revival of "Meta Days," a since-canceled holiday program that allowed staff to take a few extra days off. [1]
> In response, Bosworth said that while working at Meta isn't easy, that time should be used to build more products for its users [1]
> "stop asking about Meta Days. It's very dumb." [1]
The human workers remaining will be worked to the bone, now competing with, or at least shaded and stalked to some extent, by relentless obliging machines. Or be dispensed with.
I wonder if the US will actually be fine here. We're starting to restrict immigration (at all levels, legal, illegal, skilled, unskilled) and boomers are retiring en-masse. Will the productivity gains negate the loss of labor. Labor isn't quite fungible but we might find some equilibrium between these 2 functions.
Hey all you sillies -- multi-tasking is not a thing. It never was. Please stop pretend you can split-brain when Mother Nature did not design you for it.
Here is the key: slow down. Slow down and focus on quality. Use AI to crush one task to high quality. Pour tokens on the problem until it shines. Derive lessons from that to apply going forward.
Learning and efforts to uncover patterns always require front-loading, even when moving fast/agile/blablabla. You need to learn what a good job looks like first before you try and scale up.
Ignoring this lesson WILL cause you to burn out. The Iron Triangle remains undefeated.
996 already became the new normal, it just wasn't evenly distributed. The perceived latency was mostly in the end-output, but you can see it happening as China became dominant not through scientific breakthroughs or discoveries, but through raw unopposed effort put in. Even if two people can't do the work of one, throwing five or ten people at the problem will work.
Now with LLMs which can effectively work nonstop, the only remaining bottleneck is whoever has to be around to approve or otherwise take control.
Biggest issue that gets lost is the slower mundane parts of development were/are a very valuable forcing function to think about wtf you are even doing and making.
Even with AI, you need that time. So I've drawn a line in sand with management that, no, I won't be rushed along and hurry up on those important parts, even with AI speeding some things up.
The only consistent feature that has allowed people to work less and see more of the benefit of productivity is organized labor, and it's not even close.
It was always a means to end. You do "work" to drive impact. You didn't think your job was to write code, did you? Your job is to be a cog in a machine that crushes competition. Your job is not about you or your feelings. You can do your "job" now faster? Great, figure out a way to do it with more impact. If you can't do that, find another job. If you can do that at your current job, great, figure out a way to do that. If you don't like this setup, you can do other jobs. Play the game or get destroyed. Read a history book or look at the guy bringing your your Chinese crap from Amazon for alternatives.
This is a general trend with technology, and is why the industrial revolution never gave us an age of leisure. First people are worried it's gonna take all the jobs, then they realize no it's actually going to create a ton more jobs and more work by expanding the overall economy and creating dozens of new niches for new stuff that didn't exist before. Some old professions vanish but new ones either get created or get expanded a lot.
AI needs humans to fill in the gaps at what it's not good at: acquiring and managing large volumes of new information/context. This requires humans to learn faster than ever before, since their projects grow much more rapidly, and to spend far more time managing and switching context, which people generally find unpleasant.
This is a bit of a strawman, as almost all AI-related opinion pieces are on here lately.
"The pitch was that AI would give us our time back. Write less boilerplate, delegate the boring parts, go home earlier."
Who pitched this? Where was this argued?
And for what it's worth, our entire industry was pitched as a way for businesses to do more with less. Our raison d'être is that we would automate and enhance parts of business so you no longer need the rooms of analysts, the constant inventory manpower churn, rooms of typists, mailrooms, and on and on and on.
It's a bit rich for us to clutch pearls now. We exist as an industry to put other people out of work, and many of us have made a lot of money doing just that. There is zero surprise that businesses would like to put us out of work if they could.
And even among our industry, many of us spent much of our time trying to reduce the need for other programmers. Doesn't the website builder maker put web developers out of business? Framework vendors are replacing lots of internal code that could keep big teams employed for decades! Microsoft and other big firms are constantly trying to figure out what the many employers of software dev teams are doing...so they can replace them. I mean, we all do that, trying to automate and simplify the need for other teams to exist.
The "pitch" of AI is that it's a new tool that your competitors are going to be using, so you had better get onboard now too, in a classic adapt or die choice. And...this is largely true! You can't argue it away. You can mythologize a never existed history where software used to be perfect in the pre-AI era. You can't moralize about slop or pretend it's all a mirage. It's all just foolish rhetoric that groups all cluck and upvote to make each other feel good, while the rest of the world moves on.
I stopped reading when I got to “The twist is the role”.
I read enough clanker writing at work.
Which is too bad because I really empathize with having to juggle multiple sessions while knowing that I do my best work when I focus and now I have to do the opposite of focus.
If productivity is raised, who will profit from the surplus of said productivity increase? The natural thought throughout the centuries has been that once the machines do more work, the people who had to do the shitty labor will have more spare time to do things.
The reality is of course that we live in a capitalist system and if productivity increases there are people who will be 100% focused on ensuring those gains go towards the profits of their shareholders. Meaning a new machine doesn't mean you let your workers work less, it means you fire the workers you don't need and require the others to replace the work of the ones who are gone.
Now this reaches white collar workers who always thought themselves to be somewhat immune from that issue, but this is a centuries old tale for the blue collar folk. And from them you could also learn how to successfully deal with that.
To be clear to the author and everyone else: AI never promised less work, or a better society. Those promises were made by humans with a storied history of misdirection, misinformation, and outright lies, and those same humans have intentionally refused or failed to deliver on those promises because, well, why would they? You bought into their scheme, they got their politicians in their pockets, they have outsized influence in society, and they have blank checks to basically do as they please written by some of the worst humans society has minted.
The doom was never about the tools, it was always about the shitty humans shoving it down our throats to their own benefit.
I'm doing a lot of investigations using AI. I know what to look into, I'm just using AI to do the menial tasks (pull logs from there, see if X is happening, edit that grafana dashboard, etc etc).
And frankly I see that mostly AI-doomers voices have been heard. I have a colleague from finance dept that's ecstatic about having a claude plugin for excel. She still reviews the sheets she compiles but she's so happy to have an incredible tool to take care of the menial work (that is, a conversational interface rather than having to go click left and right to build the kind of visualization she wants etc etc).
you never worked more before because considering AI-Based coding you're extremely more efficient in every task you accomplish. Everything is solved faster, and that makes more problems to come. It's like an unevitable escalation. Such putting your pedal on full gas without knowing how to leave it from there.
I feel this. Thinking about the domain, typing a bit, maybe having a new idea while I'm eating on a break now feels like a memory of heaven. Today I'm constantly going over painful "grilling" sessions where I have to read mountains of dense prose where each word maps to some "glossary" from previously generated docs and figure out what is overengineered bullshit and what is actually necessary.
No. Actually some things take longer and lead to worse outcomes. AI tools invent problems and find work for themselves while also failing at their stated tasks in opaque ways all the damn time. It's obnoxious and why so many repos are stupidly large for no good reason.
This is what I think is the solution - I've tried it a few times with good results.
1. We need our own project or task plan. A simple forward-thinking TODO list is good enough. This needs to be prepared before the day begins. Daily.
2. Focus and limit scope. Focusing on doing ONLY what's on the list is the discipline I (we) lack and which we need badly.
3. Side quests are killing us. Allowing ourselves to veer off the planned daily tasks should require a complete away-from-the-keyboard planning session with pen and paper. Print it out but plan it away from the LLM. Use the LLM to summarise the problem, by all means, but do the thinking away from the LLM. The current TODO list needs to be the main input of such a deliberation. The output is the new TODO list. The objective of this away-from-keyboard exercise is simply to maintain control of the next steps.
On days where I worked with my TODO lists, I clocked out and joined the family for activities, slept sooner and better. On days I lost control of that list, the exhaustion mounted almost on the same day.
Make a finite list for the day, tick it off and go spend time with the missus, the kids or the dog. 2c.