> Bring up AI in a Seattle coffee shop now and people react like you're advocating asbestos.
can you please share the methodology you used to reach this conclusion?
in other words - what is the sample size? how many Seattle coffee shops did you walk into and yell out "hey, what do people think about AI?" (or did you gather the data in a different way, such as by approaching individual people at the coffee shop?)
what is your control group? in other words, how many SF coffee shops did you visit and conduct the same experiment?
Anecdotally, lots of people in SF tech hate AI too. _Most_ people out of tech do.
But, enough of the people in tech have their future tied to AI that there are lot of vocal boosters.
'If you could classify your project as "AI," you were safe and prestigious. If you couldn't, you were nobody. Overnight, most engineers got rebranded as "not AI talent."'
It hits weirdly close to home. Our leadership did not technically mandate use, but 'strongly encourages' it. I did not even have my review yet, but I know that once we get to the goals part, use of AI tools will be an actual metric ( which is.. in my head somewhere between skeptic and evangelist.. dumb ).
But the 'AI talent' part fits. For mundane stuff like data model, I need full committee approval from people, who don't get it anyway ( and whose entire contribution is: 'what other companies are doing' ).
textbook way to NOT rollout AI for your org. AI has genuine benefits to white collar workers, but they are not trained for the use-cases that would actually benefit them, nor are they trained in what the tech is actually good at. they are being punished for using the tools poorly (with no guidance on how to use them "good"), and when they use the tools well, they fear being laid off once an SOP for their AI workflows are written.
> My former coworker—the composite of three people for anonymity—now believes she's both unqualified for AI work and *that AI isn't worth doing anyway*. *She's wrong on both counts*, but the culture made sure she'd land there.
I'm not sure they're as wrong as these statements imply?
Do we think there's more or less crap out now with the advent and pervasiveness of AI? Not just from random CEOs pushing things top down, but even from ICs doing their own gig?
It's not just that AI is being pushed on to employees by the tech giants - this is true - but that the hype of AI as a life changing tech is not holding up and people within the industry can easily see this. The only life-changing thing it's doing is due to a self-fulfilling prophecy of eliminating jobs in the tech industry and outside by CEOs who have bet too much on AI. Everyone currently agrees that there is no return on all the money spent on AI. Some players may survive and do well in the future but for a majority there is only the prospect of pain, and this is what all the negativity is about.
> I wanted her take on Wanderfugl , the AI-powered map I've been building full-time.
I can at least give you one piece of advice. Before you decide on a company or product name, take the time to speak it out loud so you can get a sense of how it sounds.
The weird thing is that half of the uses of the name on that landing page spell it as "Wanderfull". All of the mock-up screencaps use it, and at the bottom with "Be one of the first people shaping Wanderfull" etc.
Literally everyone I know is sick of AI. Sick of it being crowbar'd into tools we already use and find value in. Sick of it being hyped at us as though it's a tech moment it simply isn't. Sick of companies playing at being forward thinking and new despite selling the same old shit but they've bolted a chatbot to it, so now it's "AI." Sick of integrations and products that just plain do not fucking work.
I wouldn't shit talk you to your face if you're making an AI thing. However I also understand the frustration and the exhaustion with it, and to be blunt, if a product advertises AI in it, I immediately do treat it more skeptically. If the features are opt-in, fine. If however it seems like the sort of thing that's going to start spamming me with Clippy-style "let our AI do your work for you!" popups whilst I'm trying to learn your fucking software, I will get aggravated extremely fast.
This isn’t just a Seattle thing, but I do think the outsized presence of specific employers there contributes to an outsized negativity around AI.
Look, good engineers just want to do good work. We want to use good tools to do good work, and I was an early proponent of using these tools in ways to help the business function better at PriorCo. But because I was on the wrong team (On-Prem), and because I didn’t use their chatbots constantly (I was already pitching agents before they were a defined thing, I just suck at vocabulary), I was ripe for being thrown out. That built a serious resentment towards the tooling for the actions of shitty humans.
I’m not alone in these feelings of resentment. There’s a lot of us, because instead of trusting engineers to do good work with good tools, a handful of rich fucks decided they knew technology better than the engineers building the fucking things.
"I said, Imagine how cool would this be if we had like, a 10-foot wall. It’s interactive and it’s historical. And you could talk to Martin Luther King, and you could say, ‘Well, Dr, Martin Luther King, I’ve always wanted to meet you. What was your day like today? What did you have for breakfast?’ And he comes back and he talks to you right now."
I'm stuck between feeling bad because this is my field–I spend most days worrying about not being able to pay my bills or get another job–and wanting to shake every last tech worker by the shoulders and yell "WAKE UP!" at them. If you are unhappy with what your employer is doing, because they have more power over you, you don't have to just sit there and take it. You can organize.
Of course, you could also go online and sulk, I suppose. There are more options between "ZIRP boomtimes lol jobs for everyone!" and "I got fired and replaced with ELIZA". But are tech workers willing to expore them? That's the question.
It just feels like it's in bad taste that we have the most money and privilege and employment left (despite all of the doom and gloom), and we're sitting around feeling sorry for ourselves. If not now, when? And if not us, who?
I wonder if I'm the guy in the bubble or if all these people are in the bubble. Everyone I know is really enjoying using these tools. I wrote a comment yesterday about how much my life has improved https://news.ycombinator.com/item?id=46131280
But also, it's not just my own. My wife's a graphic designer. She uses AI all the time.
Honestly, this has been revolutionary for me for getting things done.
> After a pause I tried to share how much better I've been feeling—how AI tools helped me learn faster, how much they accelerated my work on Wanderfugl. I didn't fully grok how tone deaf I was being though. She's drowning in resentment.
Here's the deal. Everyone I know who is infatuated with AI shares things AI told them with me, unsolicited, and it's always so amazingly garbage, but they don't see it or they apologize it away [1]. And this garbage is being shoved in my face from every angle --- my browser added it, my search engine added it, my desktop OS added it, my mobile OS added it, some of my banks are pushing it, AI comment slop is ruining discussion forums everywhere (even more than they already were, which is impressive!). In the mean time, AI is sucking up all the GPUs, all the RAM, and all the kWH.
If AI is actually working for you, great, but you're going to have to show it. Otherwise, I'm just going to go into my cave and come out in 5 years and hope things got better.
[1] Just a couple days ago, my spouse was complaining to her friend about a change that Facebook made, and her friend pasted an AI suggestion for how to fix it with like 7 steps that were all fabricated. That isn't helpful at all. It's even less helpful than if the friend just suggested to contact support and/or delete the facebook account.
I just picked up an old gamecube. it's refreshing to play purely offline content from an age without any AI art of any kind. some games, like animal crossing, will break in 2031 though, so there's only a good 5 more years left to enjoy it.
To the extent that Microsoft pushes their employees to use all their other shitty products, Copilot seeks like just another one (it can't be more miserable/broken than Sharepoint).
I honestly expected this to be about sanctimonious lefties complaining about a single chatgpt query using an Olympic swimming pool worth of water, but it was actually about Seattle big tech workers hating it due to layoffs and botched internal implementations which is a much more valid reason to hate it.
My buddies still or until recently still at Amazon have definitely been feeling this same push. Internal culture there has been broken since the post covid layoffs, and layering "AI" over the layoffs leaves a bad taste.
> Engineers don't try because they think they can't.
This article assumes that AI is the centre of the universe, failing to understand that that assumption is exactly what's causing the attitude they're pointing to.
There's a dichotomy in the software world between real products (which have customers and use cases and make money by giving people things they need) and hype products (which exist to get investors excited, so they'll fork over more money). This isn't a strict dichotomy; often companies with real products will mix in tidbits of hype, such as Microsoft's "pivot to AI" which is discussed in the article. But moving toward one pole moves you away from the other.
I think many engineers want to stay as far from hype-driven tech as they can. LLMs are a more substantive technology than blockchain ever was, but like blockchain, their potential has been greatly overstated. I'd rather spend my time delivering value to customers than performing "big potential" to investors.
So, no. I don't think "engineers don't try because they think they can't." I think engineers KNOW they CAN and resent being asked to look pretty and do nothing of value.
> But moving toward one pole moves you away from the other.
My assumption detector twigged at that line. I think this is just replacing the dichotomy with a continuum between two states. But the hype proponents always hope - and in some cases they are right - that those two poles overlap. People make and lose fortunes on placing those bets and you don't necessarily have to be right or wrong in an absolute sense, just long enough that someone else will take over your load and hopefully at a higher valuation.
Engineers are not usually the ones placing the bets, which is why they're trying to stay away from hype driven tech (to them it is neutral with respect to the outcome but in case of a failure they lose their job, so better to work on things that are not hyped, it is simply safer). But as soon as engineers are placing bets they are just as irrational as every other class of investor.
I've never worked at Microsoft. However, I do have some experience with the company.
I worked building tools within the Microsoft ecosystem, both on the SQL Server side, and on the .NET and developer tooling side, and I spent some time working with the NTVS team at Microsoft many years ago, as well as attending plenty of Microsoft conferences and events, working with VSIP contacts, etc. I also know plenty of people who've worked at or partnered with Microsoft.
And to me this all reads like classic Microsoft. I mean, the article even says it: whatever you're doing, it needs to align with whatever the current key strategic priority is. Today that priority is AI, 12 years ago it was Azure, and on and on. And, yes, I'd imagine having to align everything you do to a single priority regardless of how natural that alignment is (or not) gets pretty exhausting, and I'd bet it's pretty easy to burn out on it if you're in an area of the business where this is more of a drag and doesn't seem like it delivers a lot of value. And you'll have to dogfood everything (another longtime Microsoft pattern) core to that priority even if it's crap compared with whatever else might be out there.
But I don't think it's new: it's simply part and parcel of working at Microsoft. And the thing is, as a strategy it's often served them well: Windows[0], Xbox, SQL Server, Visual Studio, Azure, Sharepoint, Office, etc. Doesn't always work, of course: Windows Phone went really badly, but it's striking that this kind of swing and a miss is relatively rare in Microsoft's history.
And so now, of course, they're doing it with AI. And, of course, they're a massive company, so there will be plenty of people there who really aren't having a good time with it. But, although it's far from a foregone conclusion, it would not be a surprise for Microsoft to come from behind and win by repeating their usual strategy... again.
[0] Don't overread this: I'm not necessarily saying I'm a huge fan. In fact I do think Windows, at is core, is a decent operating system, and has been for a very long time. On the back end it works well, and I have no complaints. But I viscerally despise Windows 11 as a desktop operating system. That's right: DESPISE. VISCERALLY. AT A MOLECULAR LEVEL.
This somewhat reflects my sentiment to this article. It felt very condescending. This "self-limiting beliefs" and the implication that Seattle engineers are less than San Francisco engineers because they haven't bought into AI...well, neither have all the SF engineers.
One interesting take away from the article and the discussion is that there seem to be two kinds of engineers: those that buy into the hype and call it "AI," and those that see it for the fancy search engine it is and call it an "LLM." I'm pretty sure these days when someone mentions "AI" to me I roll my eyes. But if they say, "LLM," ok, let's have a discussion.
Ex-Google here; there are many people both current and past-Google that feel the same way as the composite coworker in the linked post.
I haven't escaped this mindset myself. I'm convinced there are a small number of places where LLMs make truly effective tools (see: generation of "must be plausible, need not be accurate" data, e.g. concept art or crowd animations in movies), a large number of places where LLMs make apparently-effective tools that have negative long-term consequences (see: anything involving learning a new skill, anything where correctness is critical), and a large number of places where LLMs are simply ineffective from the get-go but will increasingly be rammed down consumers' throats.
Accordingly I tend to be overly skeptical of AI proponents and anything touching AI. It would be nice if I was more rational, but I'm not; I want everyone working on AI and making money from AI to crash and burn hard. (See also: cryptocurrency)
My experience is the productivity gains are negative to neutral. Someone else basically wrote that the total "work" was simply being moved from one bucket to another. (I can't find the original link.)
Example: you might spend less time on initial development, but more time on code review and rework. That has been my personal experience.
Is it true that it's bad for learning new skills? My gut tells me it's useful as long as I don't use it to cheat the learning process and I mainly use it for things like follow up questions.
Not sure if that's also category #2 or a new one, but also: Places where AI is at risk of effectively becoming a drug and being actively harmful for the user: Virtual friends/spouses, delusion-confirming sycophants, etc.
I also hoped it would crash and burn. The real value added usecases will remain. The overhyped crap won't.
But the shockwave will cause a huge recession and all those investors that put up trillions will not take their losses. Rich people never get poorer. One way or another us consumers will end up paying for their mistakes. Either by huge inflation, job losses, energy costs, service enshittification whatever. We're already seeing the memory crisis having huge knock on effects with next year's phones being much more expensive. That's one of the ways we are going to be paying for this circus.
I really see value in it too, sure. But the amount of investment that goes into it is insane. It's not that valuable by far. LLMs are not good for everything and the next big thing is still a big question mark. AI is dragged in by the hair into usecases where it doesn't belong. The same shit we saw with blockchains, but now on a world crashing scale. It's very scary seeing so much insanity.
But anyway whatever I think doesn't matter. Whatever happens will happen.
My friends and I have always wondered as we've gotten older what's going to be the new tech that the younger generation seems to know and understand innately while the older generations remain clueless and always need help navigating (like computers/internet for my parents' generation and above). I am convinced that thing is AI.
Kids growing up today are using AI for everything, whether or not that's sanctioned or if it's ultimately helpful or harmful to their intellectual growth. I think the jury is still out on that. But I do remember growing up in the 90s, spending a lot of time on the computer, older people would remark how I'll have no social skills, I won't be able to write cursive or do arithmetic in my head, won't learn any real skills, etc, turns out I did just fine and now those same people always have to call me for help when they run into the smallest issue with technology.
I think a lot of people here are going to become roadkill if they refuse to learn how to use these new tools. I just built a web app in 3 weeks with only prompts to Claude Code, I didn't write a single line of code, and it works great. It's pretty basic, but probably would have taken me 3+ months instead of 3 weeks doing it the old fashioned way. If you tried it once a year ago and have written it off, a lot has changed since then and the tools continue to improve every month. I really think that eventually no one will be checking code just like hardly anyone checks the assembly output of a compiler anymore.
You have to understand how the context window works, how to establish guardrails so you're not wasting time repeating the same things over and over again, force it to check its own work with lots of tests, etc. It's really a game changer when you can just say in one prompt "write me an admin dashboard that displays users, sessions, and orders with a table and chart going back 30 days" or "wire up my site for google analytics, my tag code is XXXXXXX" and it just works.
I basically agree. OK: Small focused models for specific use cases, small models like the new mistral-3-3B that I found today to be good at tool use I and thus for building narrow ranged applications.
I have been mostly been paid to work on AI projects since 1982, but I want to pull my hair out and scream over the big push in the USA to develop super-AGI. Such a waste of resources and such a hit on society that needs resources used for better purposes.
> [...] a large number of places where LLMs make apparently-effective tools that have negative long-term consequences (see: anything involving learning a new skill, [...]
Don't people learn from imperfect teachers all the time?
As a gamedev, there's nothing I hate more than AI concept art. It's always soulless. The best thing about games is there's no limit to human imagination, and you can make whatever you want. But when we leave the imagination stage to a computer then leave the final brushing up to humans, we're getting the order completely backwards. It's bonkers and just disgusting to me.
That said, game engine documentation is often pretty hard to navigate. Most of the best information is some YouTube video recorded by some savant 15 year old with a busted microphone. And you need to skim through 30 minutes of video until you find what you need. The biggest problem is not knowing what you don't know, so it's hard to know where to begin. There are a lot of things you may think you need to spend 2 days implementing, but the engine may have a single function and a couple built in settings to do it.
Where LLMs shine is that I can ask a dumb question about this stuff, and can be pointed in the right direction pretty quickly. The implementation it spits out is often awful (if not unusable), but I can ask a question and it'll name drop the specific function and setting names that'll save me a lot of work. And from there, I know what to look up and it's a clear path from there.
And gamedev is a very strong case of not needing a correct solution. You just need things to feel right for most cases. Games that are rough around the edges have character. So LLM assistance for implementation (not art) can be handy.
This includes IME the initial stages of art creation (the planning, not generating, stage). It's kind of like having someone to bounce ideas off of at 3am. It's a convenient way of trigging your own brain to be inspired.
Thanks for the post - it's work to write and synthesize, and I always appreciate it!
My first reaction was "replace 'AI' with the word 'Cloud'" ca 2012 at MS; what's novel here?
With that in mind, I'm not sure there is anything novel about how your friend is feeling or the organizational dynamics, or in fact how large corporations go after business opportunities; on those terms, I think your friends' feelings are a little boring, or at least don't give us any new market data.
In MS in that era, there was a massive gold rush inside the org to Cloud-ify everything and move to Azure - people who did well at that prospered, people who did not, ... often did not. This sort of internal marketplace is endemic, and probably a good thing at large tech companies - from the senior leadership side, seeing how employees vote with their feet is valuable - as is, often, the directional leadership you get from a Satya who has MUCH more information than someone on the ground in any mid-level role.
While I'm sure there were many naysayers about the Cloud in 2012, they were wrong, full stop. Azure is immensely valuable. It was right to dig in on it and compete with AWS.
I personally think Satya's got a really interesting hyper scaling strategy right now -- build out national-security-friendly datacenters all over the world -- and I think that's going to pay -- but I could be wrong, and his strategy might be much more sophisticated and diverse than that; either way, I'm pretty sure Seattleites who hate how AI has disrupted their orgs and changed power politics and winners and losers in-house will have to roll with the program over the next five years and figure out where they stand and what they want to work on.
1. You were a therapy session for her. Her negativity was about the layoffs.
2. FAANG companies dramatically overhired for years and are using AI as an excuse for layoffs.
3. AI scene in Seattle is pretty good, but as with everywhere else was/is a victim of the AI hype. I see estimates of the hype being dead in a year. AI won't be dead, but throwing money at the whatever Uber-for-pets-AI-ly idea pops up won't happen.
4. I don't think people hate AI, they hate the hype.
Anyways, your app actually does sound interesting so I signed up for it.
I was an early employee at a unicorn and I saw this culture take hold once we started hiring from Big Tech talent pools and offering Big Tech comp packages, though before AI hype took off. There's a crazy lack of agency that kicks in for Big Tech folks that's really hard to explain. This feeling that each engineer is this mercenary trying really hard to not get screwed by the internal system.
Most of it is because there's little that ties actual output to organizational outcomes. AI mandates after all are just a way to bluntly for e engineers to use AI, where if you were at a startup or smaller company you would probably organically find how much an LLM helps you where. It may not even help your actual work even if it helps your coworkers. That market feedback is sorely missing from the Big Techs and so hamfisted engineering mandates have to do in order to for e engineers to become more efficient.
In these cases I always try to remind friends that you can always leave a Big Tech. The thing is, from what I can tell, a lot of these folks have developed lifestyle inflation from working in Big Tech and some of their anger comes from feeling trapped in their Big Tech role due to this. While I understand, I'm not particularly sympathetic to this viewpoint. At the end of the day your lifestyle is in your hands.
Thanks for signing up. I’m going to try really hard to open up some beta slots next week so more people can try it. There’s some embarrassingly bad bugs in prod right now…
It's the closing trash compactor of soullessness and hate of the human, described vividly as having affected Microsoft culture as thoroughly as intergranular corrosion can turn a solid block of aluminum to dust.
Fuck Microsoft for both hating me and hating their own people. Fuck. That. Shit.
There's a great non-AI point in this article - Seattle has great engineers. In pursuing startups, Seattle engineers are relatively unambitious compared to the Bay Area. By that I mean there's less "shooting for unicorns" and a comparatively more reserved startup culture and environment.
I'm not sure why. I don't think it's access to capital, but I'd love to hear thoughts.
236 comments
[ 2.6 ms ] story [ 140 ms ] thread> Bring up AI in a Seattle coffee shop now and people react like you're advocating asbestos.
can you please share the methodology you used to reach this conclusion?
in other words - what is the sample size? how many Seattle coffee shops did you walk into and yell out "hey, what do people think about AI?" (or did you gather the data in a different way, such as by approaching individual people at the coffee shop?)
what is your control group? in other words, how many SF coffee shops did you visit and conduct the same experiment?
It hits weirdly close to home. Our leadership did not technically mandate use, but 'strongly encourages' it. I did not even have my review yet, but I know that once we get to the goals part, use of AI tools will be an actual metric ( which is.. in my head somewhere between skeptic and evangelist.. dumb ).
But the 'AI talent' part fits. For mundane stuff like data model, I need full committee approval from people, who don't get it anyway ( and whose entire contribution is: 'what other companies are doing' ).
I'm not sure they're as wrong as these statements imply?
Do we think there's more or less crap out now with the advent and pervasiveness of AI? Not just from random CEOs pushing things top down, but even from ICs doing their own gig?
"I made this half-pony, half-monkey monster to please you
But I get the feeling that you don't like it
What's with all the screaming?
You like monkeys, you like ponies
Maybe you don't like monsters so much
Maybe I used too many monkeys
Isn't it enough to know that I ruined a pony
Making a gift for you?
> I wanted her take on Wanderfugl , the AI-powered map I've been building full-time.
I can at least give you one piece of advice. Before you decide on a company or product name, take the time to speak it out loud so you can get a sense of how it sounds.
So even the creator can't decide what to call it!
this must be one of the incredible AI innovations the folks in Seattle are missing out on
I wouldn't shit talk you to your face if you're making an AI thing. However I also understand the frustration and the exhaustion with it, and to be blunt, if a product advertises AI in it, I immediately do treat it more skeptically. If the features are opt-in, fine. If however it seems like the sort of thing that's going to start spamming me with Clippy-style "let our AI do your work for you!" popups whilst I'm trying to learn your fucking software, I will get aggravated extremely fast.
Look, good engineers just want to do good work. We want to use good tools to do good work, and I was an early proponent of using these tools in ways to help the business function better at PriorCo. But because I was on the wrong team (On-Prem), and because I didn’t use their chatbots constantly (I was already pitching agents before they were a defined thing, I just suck at vocabulary), I was ripe for being thrown out. That built a serious resentment towards the tooling for the actions of shitty humans.
I’m not alone in these feelings of resentment. There’s a lot of us, because instead of trusting engineers to do good work with good tools, a handful of rich fucks decided they knew technology better than the engineers building the fucking things.
"I said, Imagine how cool would this be if we had like, a 10-foot wall. It’s interactive and it’s historical. And you could talk to Martin Luther King, and you could say, ‘Well, Dr, Martin Luther King, I’ve always wanted to meet you. What was your day like today? What did you have for breakfast?’ And he comes back and he talks to you right now."
Of course, you could also go online and sulk, I suppose. There are more options between "ZIRP boomtimes lol jobs for everyone!" and "I got fired and replaced with ELIZA". But are tech workers willing to expore them? That's the question.
It just feels like it's in bad taste that we have the most money and privilege and employment left (despite all of the doom and gloom), and we're sitting around feeling sorry for ourselves. If not now, when? And if not us, who?
But also, it's not just my own. My wife's a graphic designer. She uses AI all the time.
Honestly, this has been revolutionary for me for getting things done.
Here's the deal. Everyone I know who is infatuated with AI shares things AI told them with me, unsolicited, and it's always so amazingly garbage, but they don't see it or they apologize it away [1]. And this garbage is being shoved in my face from every angle --- my browser added it, my search engine added it, my desktop OS added it, my mobile OS added it, some of my banks are pushing it, AI comment slop is ruining discussion forums everywhere (even more than they already were, which is impressive!). In the mean time, AI is sucking up all the GPUs, all the RAM, and all the kWH.
If AI is actually working for you, great, but you're going to have to show it. Otherwise, I'm just going to go into my cave and come out in 5 years and hope things got better.
[1] Just a couple days ago, my spouse was complaining to her friend about a change that Facebook made, and her friend pasted an AI suggestion for how to fix it with like 7 steps that were all fabricated. That isn't helpful at all. It's even less helpful than if the friend just suggested to contact support and/or delete the facebook account.
To be fair, pretty much all advice in life is less helpful than 'delete the facebook account'
I just picked up an old gamecube. it's refreshing to play purely offline content from an age without any AI art of any kind. some games, like animal crossing, will break in 2031 though, so there's only a good 5 more years left to enjoy it.
My buddies still or until recently still at Amazon have definitely been feeling this same push. Internal culture there has been broken since the post covid layoffs, and layering "AI" over the layoffs leaves a bad taste.
This article assumes that AI is the centre of the universe, failing to understand that that assumption is exactly what's causing the attitude they're pointing to.
There's a dichotomy in the software world between real products (which have customers and use cases and make money by giving people things they need) and hype products (which exist to get investors excited, so they'll fork over more money). This isn't a strict dichotomy; often companies with real products will mix in tidbits of hype, such as Microsoft's "pivot to AI" which is discussed in the article. But moving toward one pole moves you away from the other.
I think many engineers want to stay as far from hype-driven tech as they can. LLMs are a more substantive technology than blockchain ever was, but like blockchain, their potential has been greatly overstated. I'd rather spend my time delivering value to customers than performing "big potential" to investors.
So, no. I don't think "engineers don't try because they think they can't." I think engineers KNOW they CAN and resent being asked to look pretty and do nothing of value.
The wealthiest person in the world relies entirely on his ability to convince people to accept hype that surpasses all reason.
My assumption detector twigged at that line. I think this is just replacing the dichotomy with a continuum between two states. But the hype proponents always hope - and in some cases they are right - that those two poles overlap. People make and lose fortunes on placing those bets and you don't necessarily have to be right or wrong in an absolute sense, just long enough that someone else will take over your load and hopefully at a higher valuation.
Engineers are not usually the ones placing the bets, which is why they're trying to stay away from hype driven tech (to them it is neutral with respect to the outcome but in case of a failure they lose their job, so better to work on things that are not hyped, it is simply safer). But as soon as engineers are placing bets they are just as irrational as every other class of investor.
I worked building tools within the Microsoft ecosystem, both on the SQL Server side, and on the .NET and developer tooling side, and I spent some time working with the NTVS team at Microsoft many years ago, as well as attending plenty of Microsoft conferences and events, working with VSIP contacts, etc. I also know plenty of people who've worked at or partnered with Microsoft.
And to me this all reads like classic Microsoft. I mean, the article even says it: whatever you're doing, it needs to align with whatever the current key strategic priority is. Today that priority is AI, 12 years ago it was Azure, and on and on. And, yes, I'd imagine having to align everything you do to a single priority regardless of how natural that alignment is (or not) gets pretty exhausting, and I'd bet it's pretty easy to burn out on it if you're in an area of the business where this is more of a drag and doesn't seem like it delivers a lot of value. And you'll have to dogfood everything (another longtime Microsoft pattern) core to that priority even if it's crap compared with whatever else might be out there.
But I don't think it's new: it's simply part and parcel of working at Microsoft. And the thing is, as a strategy it's often served them well: Windows[0], Xbox, SQL Server, Visual Studio, Azure, Sharepoint, Office, etc. Doesn't always work, of course: Windows Phone went really badly, but it's striking that this kind of swing and a miss is relatively rare in Microsoft's history.
And so now, of course, they're doing it with AI. And, of course, they're a massive company, so there will be plenty of people there who really aren't having a good time with it. But, although it's far from a foregone conclusion, it would not be a surprise for Microsoft to come from behind and win by repeating their usual strategy... again.
[0] Don't overread this: I'm not necessarily saying I'm a huge fan. In fact I do think Windows, at is core, is a decent operating system, and has been for a very long time. On the back end it works well, and I have no complaints. But I viscerally despise Windows 11 as a desktop operating system. That's right: DESPISE. VISCERALLY. AT A MOLECULAR LEVEL.
One interesting take away from the article and the discussion is that there seem to be two kinds of engineers: those that buy into the hype and call it "AI," and those that see it for the fancy search engine it is and call it an "LLM." I'm pretty sure these days when someone mentions "AI" to me I roll my eyes. But if they say, "LLM," ok, let's have a discussion.
I haven't escaped this mindset myself. I'm convinced there are a small number of places where LLMs make truly effective tools (see: generation of "must be plausible, need not be accurate" data, e.g. concept art or crowd animations in movies), a large number of places where LLMs make apparently-effective tools that have negative long-term consequences (see: anything involving learning a new skill, anything where correctness is critical), and a large number of places where LLMs are simply ineffective from the get-go but will increasingly be rammed down consumers' throats.
Accordingly I tend to be overly skeptical of AI proponents and anything touching AI. It would be nice if I was more rational, but I'm not; I want everyone working on AI and making money from AI to crash and burn hard. (See also: cryptocurrency)
Example: you might spend less time on initial development, but more time on code review and rework. That has been my personal experience.
But the shockwave will cause a huge recession and all those investors that put up trillions will not take their losses. Rich people never get poorer. One way or another us consumers will end up paying for their mistakes. Either by huge inflation, job losses, energy costs, service enshittification whatever. We're already seeing the memory crisis having huge knock on effects with next year's phones being much more expensive. That's one of the ways we are going to be paying for this circus.
I really see value in it too, sure. But the amount of investment that goes into it is insane. It's not that valuable by far. LLMs are not good for everything and the next big thing is still a big question mark. AI is dragged in by the hair into usecases where it doesn't belong. The same shit we saw with blockchains, but now on a world crashing scale. It's very scary seeing so much insanity.
But anyway whatever I think doesn't matter. Whatever happens will happen.
Kids growing up today are using AI for everything, whether or not that's sanctioned or if it's ultimately helpful or harmful to their intellectual growth. I think the jury is still out on that. But I do remember growing up in the 90s, spending a lot of time on the computer, older people would remark how I'll have no social skills, I won't be able to write cursive or do arithmetic in my head, won't learn any real skills, etc, turns out I did just fine and now those same people always have to call me for help when they run into the smallest issue with technology.
I think a lot of people here are going to become roadkill if they refuse to learn how to use these new tools. I just built a web app in 3 weeks with only prompts to Claude Code, I didn't write a single line of code, and it works great. It's pretty basic, but probably would have taken me 3+ months instead of 3 weeks doing it the old fashioned way. If you tried it once a year ago and have written it off, a lot has changed since then and the tools continue to improve every month. I really think that eventually no one will be checking code just like hardly anyone checks the assembly output of a compiler anymore.
You have to understand how the context window works, how to establish guardrails so you're not wasting time repeating the same things over and over again, force it to check its own work with lots of tests, etc. It's really a game changer when you can just say in one prompt "write me an admin dashboard that displays users, sessions, and orders with a table and chart going back 30 days" or "wire up my site for google analytics, my tag code is XXXXXXX" and it just works.
I have been mostly been paid to work on AI projects since 1982, but I want to pull my hair out and scream over the big push in the USA to develop super-AGI. Such a waste of resources and such a hit on society that needs resources used for better purposes.
Don't people learn from imperfect teachers all the time?
That said, game engine documentation is often pretty hard to navigate. Most of the best information is some YouTube video recorded by some savant 15 year old with a busted microphone. And you need to skim through 30 minutes of video until you find what you need. The biggest problem is not knowing what you don't know, so it's hard to know where to begin. There are a lot of things you may think you need to spend 2 days implementing, but the engine may have a single function and a couple built in settings to do it.
Where LLMs shine is that I can ask a dumb question about this stuff, and can be pointed in the right direction pretty quickly. The implementation it spits out is often awful (if not unusable), but I can ask a question and it'll name drop the specific function and setting names that'll save me a lot of work. And from there, I know what to look up and it's a clear path from there.
And gamedev is a very strong case of not needing a correct solution. You just need things to feel right for most cases. Games that are rough around the edges have character. So LLM assistance for implementation (not art) can be handy.
This includes IME the initial stages of art creation (the planning, not generating, stage). It's kind of like having someone to bounce ideas off of at 3am. It's a convenient way of trigging your own brain to be inspired.
My first reaction was "replace 'AI' with the word 'Cloud'" ca 2012 at MS; what's novel here?
With that in mind, I'm not sure there is anything novel about how your friend is feeling or the organizational dynamics, or in fact how large corporations go after business opportunities; on those terms, I think your friends' feelings are a little boring, or at least don't give us any new market data.
In MS in that era, there was a massive gold rush inside the org to Cloud-ify everything and move to Azure - people who did well at that prospered, people who did not, ... often did not. This sort of internal marketplace is endemic, and probably a good thing at large tech companies - from the senior leadership side, seeing how employees vote with their feet is valuable - as is, often, the directional leadership you get from a Satya who has MUCH more information than someone on the ground in any mid-level role.
While I'm sure there were many naysayers about the Cloud in 2012, they were wrong, full stop. Azure is immensely valuable. It was right to dig in on it and compete with AWS.
I personally think Satya's got a really interesting hyper scaling strategy right now -- build out national-security-friendly datacenters all over the world -- and I think that's going to pay -- but I could be wrong, and his strategy might be much more sophisticated and diverse than that; either way, I'm pretty sure Seattleites who hate how AI has disrupted their orgs and changed power politics and winners and losers in-house will have to roll with the program over the next five years and figure out where they stand and what they want to work on.
1. You were a therapy session for her. Her negativity was about the layoffs.
2. FAANG companies dramatically overhired for years and are using AI as an excuse for layoffs.
3. AI scene in Seattle is pretty good, but as with everywhere else was/is a victim of the AI hype. I see estimates of the hype being dead in a year. AI won't be dead, but throwing money at the whatever Uber-for-pets-AI-ly idea pops up won't happen.
4. I don't think people hate AI, they hate the hype.
Anyways, your app actually does sound interesting so I signed up for it.
What about the complete lack of morality some (most?) AI companies exhibit?
What about the consequences in the environment?
What about the enshitification of products?
What about the usage of water and energy?
Etc.
Most of it is because there's little that ties actual output to organizational outcomes. AI mandates after all are just a way to bluntly for e engineers to use AI, where if you were at a startup or smaller company you would probably organically find how much an LLM helps you where. It may not even help your actual work even if it helps your coworkers. That market feedback is sorely missing from the Big Techs and so hamfisted engineering mandates have to do in order to for e engineers to become more efficient.
In these cases I always try to remind friends that you can always leave a Big Tech. The thing is, from what I can tell, a lot of these folks have developed lifestyle inflation from working in Big Tech and some of their anger comes from feeling trapped in their Big Tech role due to this. While I understand, I'm not particularly sympathetic to this viewpoint. At the end of the day your lifestyle is in your hands.
I think there is no "her", the article ends with saying:
> My former coworker—the composite of three people for anonymity—now believes she's [...]
I think it's just 3 different people and they made up a "she" single coworker as a kind of example person.
I don't know, that's my reading at least, maybe I got it wrong.
Close. We're in a recession and they are using AI as an excuse for another wave of outsourcing.
>I don't think people hate AI, they hate the hype.
I hate the grift. I hate having it forced on me after refusing multiple times. That's pretty much 90% of AI right now.
It's the closing trash compactor of soullessness and hate of the human, described vividly as having affected Microsoft culture as thoroughly as intergranular corrosion can turn a solid block of aluminum to dust.
Fuck Microsoft for both hating me and hating their own people. Fuck. That. Shit.
except a lot of people really do hate AI
I'm not sure why. I don't think it's access to capital, but I'd love to hear thoughts.
I'm being course, but like... it is though.