LLMs and their capabilities are very impressive and definitely useful. The productivity gains often seem to be smaller than intuitively expected though. For example, using ChatGPT to get a response to a random question like "How do I do XYZ" is much more convenient than googling it, but the time savings are often not that relevant for your overall productivity. Before LLMs you were usually already able to find the information quickly and even a 10x speed up does not really have too much of an impact on your overall productivity, because the time it took was already negligible.
palmotea 3 days ago [-]
> For example, using ChatGPT to get a response to a random question like "How do I do XYZ" is much more convenient than googling it, but the time savings are often not that relevant for your overall productivity. Before LLMs you were usually already able to find the information quickly and even a 10x speed up does not really have too much of an impact on your overall productivity, because the time it took was already negligible.
I'd even question that. The pre-LLM solutions were in most cases better. Searching a maintained database of curated and checked information is far better than LLM output (which is possibly bullshit).
Ditto to software engineering. In software, we have things call libraries: you write the code once, test it, then you trust it and can use it as many times as you want forever for free. Why use LLM generated code when you have a library? And if you're asking for anything complex, you're probably just getting a plagiarized and bastardized version of some library anyway.
The only thing where LLMs shine is a kind of simple, lazy "mash this up so I don't have to think about it" cases. And sometimes it might be better to just do it yourself and develop your own skills instead of use an LLM.
neilalexander 2 days ago [-]
One advantage with LLMs is that they are often more able to find things that you can roughly explain but don't know the name of. They can be a good take-off point for wider searches.
davkan 13 hours ago [-]
I was able to find a decade old video via llm with the prompt “YouTube video of a french band on a radio station with a girl wearing orange jumpsuit”. I had tried many google searches without success trying to remember the video but the llm came right out with the correct video of The Dø on KEXP first try. 99 times out of 100 I prefer normal search though.
mountainriver 2 days ago [-]
Why not have the LLM generate the library?
javcasas 2 days ago [-]
Because every time you run the LLM it will generate a new library, with new and surprising bugs.
It's better to take an existing, already curated and tested library. Which, yes, may have been generated by an LLM, but has been curated beyond the skill of the LLM.
techblueberry 2 days ago [-]
The value of a library is not in the code it’s in the operations. It’s been curated and tested by multiple people in multiple environments. One of the ways to code at a high level is to delegate the cognitive load.
If you really can one shot it and it’s simple(left-pad). Great. But most things aren’t my that simple, the third time you have to think about it, it’s probably a net loss.
drzaiusx11 3 days ago [-]
If only search engine AI output didn't constantly haluciate nonexistent APIs, it might be a net productivity gain for me...but it's not. I've been bit enough times by their false "example" output for it to be a significant net time loss vs using traditional search results.
danudey 2 days ago [-]
Gemini hallucinated a method on a rust crate that it was trying to use and then spent ten minutes googling 'method_name v4l2 examples' and so on. That method doesn't exist and has never existed; there was a property on the object that contained the information it wanted, but it just sat there spinning its wheels convinced that this imagined method was the key to its success.
Eventually it gave up and commented out all the code it was trying to make work. Took me less than two minutes to figure out the solution using only my IDE's autocomplete.
It did save me time overall, but it's definitely not the panacea that people seem to think it is and it definitely has hiccups that will derail your productivity if you trust it too much.
FireBeyond 1 days ago [-]
My favorite with ChatGPT is:
"Tell me how to do X" (where X was, for one recent example, creating a Salt stanza to install and configure a service).
I do as it tells me, which seems reasonable on the face of it. But it generates an error.
"When creating X as you described, I get error: Z. Why?"
"You're absolutely correct and you should expect this error because X won't work this way. Do Y instead."
Gah... "I told you to do X, and then I'm going to act like it's not a surprise that X doesn't work and you should do something else."
dietr1ch 1 days ago [-]
You're absolutely right
Lapsa 16 hours ago [-]
it's not just that you are absolutely correct but you are also absolutely right
gloosx 23 hours ago [-]
It's even worse when LLM eats documentation for multiple versions of the same library and starts hallucimixing methods from all versions at the same time. Certainly unusable for some libraries which had a big API transition between versions recently.
skybrian 3 days ago [-]
Using ChatGPT and phrasing it like a search seems like a better way? “Can you find documentation about an API that does X?”
lazide 3 days ago [-]
It will often literally just make up the documentation.
If you ask for a link, it may hallucinate the link.
And unlike a search engine where someone had to previously think of, and then make some page with the fake content on it, it will happily make it up on the fly so you'll end up with a new/unique bit of fake documentation/url!
At that point, you would have been way better off just... using a search engine?
taude 3 days ago [-]
how is it hallucinating links? The links are direct links to the webpage that they vectorized or whatever as input to the LLM query. In fact, on almost all LLM responses DuckDuckGo and Google, the links are right there as sited sources that you click on (i know because I'm almost always clicking on the source link to read the original details, and not the made up one
madcaptenor 3 days ago [-]
I would imagine links can be hallucinated because the original URLs in the training data get broken up into tokens - so it's not hard to come up with a URL that has the right format (say https://arxiv.org/abs/2512.01234 - which is a real paper but I just made up that URL) and a plausible-sounding title.
jjj123 3 days ago [-]
Yeah, but the current state of ChatGPT doesn’t really do this. The comment you’re replying to explains why URLs from ChatGPT generally aren’t constructed from raw tokens.
madcaptenor 3 days ago [-]
You are absolutely right! The current state of ChatGPT was not in my training data.
1718627440 2 days ago [-]
How do you explain it then, when it spits out the link, that looks like it surprisingly contains the subject of your question in the URL, but that page simply doesn't exist and there isn't even a blog under that domain at all?
lazide 2 days ago [-]
Near as I can tell, people just don’t actually check and go off what it looks like it’s doing. Or they got lucky, and when they did check once it was right. Then assume it will always right.
Which would certainly explain things like hallucinated references in legal docs and papers!
The reality is that for a human to make up that much bullshit requires a decent amount of work, so most humans don’t do it - or can’t do it as convincingly. LLMs can generate nigh infinite amounts of bullshit for cheap (and often more convincing sounding bullshit than a human can do on their own without a lot of work!), making them perfect for fooling people.
Unless someone is really good at double checking things, it’s a recipe for disaster. Even worse, doing the right amount of double checking makings them often even more exhausting than just doing the work yourself in the first place.
strange_quark 2 days ago [-]
I’ve used Claude code to debug and sometimes it’ll say it knows what the issue is, then when I make it cite a source for its assertions, it will do a web search and sometimes spit out a link whose contents contradict its own claim.
One time I tried to use Gemini to figure out 1950s construction techniques so I could understand how my house was built. It made a dubious sounding claim about the foundation, so I had it give me links and keywords so I could find some primary sources myself. I was unable to find anything to back up what it told me, and then it doubled down and told me that either I was googling wrong or that what it told me was a historical “hack” that wouldn’t have been documented.
These were both recent and with the latest models, so maybe they don’t fully fabricate links, but they do hallucinate the contents frequently.
exmadscientist 2 days ago [-]
> maybe they don’t fully fabricate links
Grok certainly will (at least as of a couple months ago). And they weren't just stale links either.
lovich 2 days ago [-]
After getting beaten for telling the truth so frequently, who wouldn’t start lying?
skybrian 3 days ago [-]
I haven't seen this happen in ChatGPT thinking mode. It actually does a bunch of web searches and links to the results.
CyberDildonics 2 days ago [-]
The real benefit to a search engine is to rework and launder other people's information and make it your information.
Now instead of the wikipedia article you are reading the exact same thing from google's home page and you don't click on anything.
SkiFire13 2 days ago [-]
I think you're underestimating how many people don't know how to properly search on google (i.e. finding the proper keywords, selecting the reputable results, etc etc). Those are probably also the same people that will blindly believe anything a LLM says unfortunately.
1718627440 2 days ago [-]
True, I do not know how two properly search something on google.com in 2025. I only know how to do it on startpage.com in 2025, kagi.com in 2025 or google.com in 2015.
zelos 2 days ago [-]
LLM output is quickly rendering google search unusable, so it's kind of creating its own speedup multiplier.
robofanatic 2 days ago [-]
It really depends on what 'XYZ' is and how many hoops you need to jump through to get to the answer. ChatGPT gets information from various places and gives you the answer as well as the explanation at each step. Without tools like ChatGPT its definitely not negligible in a lot of cases.
skybrian 3 days ago [-]
I use ChatGPT “thinking” mode as a way to run multiple searches and summarize the results. It takes some time, but I can do other stuff in another tab and come back.
It’s for queries that are unlikely to be satisfied in a single search. I don’t think it would be a negligible amount of time if you did it yourself.
Incipient 2 days ago [-]
But for large searches, I then have to spend a lot of time validating the output - which I'd normally do while reading the content etc as I searched (discarding dodgy websites etc).
On the other hand, where I think llms are going to excel, is you roll the dice, trust the output, and don't validate it. If it works out yayy you're ahead of everyone else that did bother to validate it.
I think this is how vibe coded apps are going to go. If the app blows up, shut down the company and start a new one.
entropicdrifter 2 days ago [-]
I find Gemini to be the most consistent at actually using the search results for this, in "Deep Research" mode
Gud 3 days ago [-]
This is the way. I do it the same way for development. The main point is I can run multiple tasks in parallel(myself + LLM(s)).
I let Claude and ChatGPT type out code for me, while I focus on my research
direwolf20 3 days ago [-]
This is partly because Google is past the enshittification hump and ChatGPT is just starting to climb up it - they just announced ads.
hodgesrm 3 days ago [-]
This. And the wonderful thing about LLMs is that they can be trained to bend responses in specific directions, say toward using Oracle Cloud solutions. There's fertile ground for commercial value extraction that goes far beyond ads. Think of it as product placement on steroid.
direwolf20 3 days ago [-]
You don't even need training — you can add steering vectors in the middle of the otherwise-unmodified computation. Remember Golden Gate Claude?
robofanatic 2 days ago [-]
> they just announced ads
wondering how is it going to work when they "search the web" to get the information, are they essentially going to take ad revenue away from the source website?
foobarchu 2 days ago [-]
Not to be a dick, but enshittification is not a hump you get past, it's a constant climb until the product is abandoned. Did you just mean growing pains?
binary132 3 days ago [-]
The difference is that in the past that information had to come from what people wrote and are writing about, and now it can come from a derivative of an archive of what people once wrote, upon a time. So if they just stop doing that — whether because they must, or because they no longer have any reason to, or because they are now drowned out in a massive ocean of slop, or simply because they themselves have turned into slopslaves — no new information will be generated, only derivative slop, milled from derivative slop.
I think we all understand that at this point, so I question deeply why anyone acts like they don’t.
sylware 3 days ago [-]
That makes me think about the development of much software out there: the development time is often several orders of magnitude smaller than its life cycle.
HarHarVeryFunny 3 days ago [-]
> For example, using ChatGPT to get a response to a random question like "How do I do XYZ" is much more convenient than googling it
More convenient than traditional search? Maybe. Quicker than traditional search? Maybe not.
Asking random questions is exactly where you run into time-wasting hallucinations since the models don't seem to be very good at deciding when to use a search tool and when just to rely on their training data.
For example, just now I was asking Gemini how to fix a bunch of Ubuntu/Xfce annoyances after a major upgrade, and it was a very mixed bag. One example: the default date and time display is in an unreadably small "date stacked over time" format (using a few pixel high font so this fits into the menu bar), and Gemini's advice was to enable the "Display date and time on single line" option ... but there is no such option (it just hallucinated it), and it also hallucinated a bunch of other suggestions until I finally figured out what you need to do is to configure it to display "Time only" rather than "Data and Time", then change the "Time" format to display both data and time! Just to experiment, I then told Gemini about this fix and amusingly the response was basically "Good to know - this'll be useful for anyone reading this later"!
More examples, from yesterday (these are not rare exceptions):
1) I asked Gemini (generally considered one of the smartest models - better than ChatGPT, and rapidly taking away market share from it - 20% shift in last month or so) to look at the GitHub codebase for an Anthropic optimization challenge, to summarize and discuss etc, and it appeared to have looked at the codebase until I got more into the weeds and was questioning it where it got certain details from (what file), and it became apparent it had some (search based?) knowledge of the problem, but seemingly hadn't actually looked at it (wasn't able to?).
2) I was asking Gemini about chemically fingerprinting (via impurities, isotopes) roman silver coins to the mines that produced the silver, and it confidently (as always) comes up with a bunch of academic references that it claimed made the connection, but none or references (which did at least exist) actually contained what it claimed (just partial information), and when I pointed this out it just kept throwing out different references.
So, it's convenient to be able to chat with your "search engine" to drill down and clarify, etc, but a big time waste if a lot of it is hallucination.
Search vs Chat has anyways really become a difference without a difference since Google now gives you the "AI Overview" (a diving off point into "AI Mode"), or you can just click on "AI Mode" in the first place - which is Gemini.
2 days ago [-]
NoGravitas 3 days ago [-]
Another reason search vs chat has become a difference without a difference is that search results are full of highly-ranked AI slop. I was searching yesterday for a way to get a Gnome-style hot corner in Windows 11, and the top result falsely asserted that hot corners were a built-in feature, and pointed to non-existing settings to enable them.
fragmede 2 days ago [-]
> I asked Gemini (generally considered one of the smartest models
Everyone is entitled to their own opinion, but I asked ChatGPT and Claude your XFCE question, and they both gave better answers than Gemini did (imo). Why would you blindly believe what someone else tells you over what you observe with your own eyes?
HarHarVeryFunny 2 days ago [-]
I'm curious what was your Claude prompt? I used to use Claude a lot more, but the free tier usage limits are very low if you use it for coding.
linuxftw 3 days ago [-]
You're overestimating the mean person's ability to search the web effectively.
jgalt212 3 days ago [-]
And perhaps both are overestimating the mean person's ability to detect a hallucinated solution vs a genuine one.
linuxftw 3 days ago [-]
I think hallucination is grossly overstated as a problem at this point, most models will actively search the web and reason about the results. You're much more likely to get the incorrect solution browsing stack overflow than you are asking AI.
danudey 2 days ago [-]
Gemini hallucinated a method name in a rust crate then spent several minutes googling the method name + 'rust example' trying to find documentation about the method it made up. Unsurprisingly it didn't find any, and then it just gave up and commented out the entire function and called it done.
linuxftw 2 days ago [-]
Comparing the free tier of Gemini to the latest premium coding models will give you drastically different results.
avaer 3 days ago [-]
The difference is LLMs let you "run Google" on your own data with copy paste. Which you could not do before.
If you're using ChatGPT like you use Google then I agree with you. But IMO comparing ChatGPT to Google means you haven't had the "aha" moment yet.
As a concrete example, a lot of my work these days involves asking ChatGPT to produce me an obscure micro-app to process my custom data. Which it usually does and renders in one shot. This app could not exist before I asked for it. The productivity gains over coding this myself are immense. And the experience is nothing like using Google.
MadDemon 3 days ago [-]
It's great for you that you were able to create this app that wouldn't otherwise exist, but does that app dramatically increase your overall productivity? And can you imagine that a significant chunk of the population would experience a similar productivity boost? I'm not saying that there is no productivity gain, but big tech has promised MASSIVE productivity gains. I just feel like the productivity gains are more modest for now, similar to other technologies. Maybe one day AGI comes along and changes everything, but I feel like we'll need a few more break throughs before that.
bryanrasmussen 3 days ago [-]
there have been various solutions that allow you to "run Google" on your own data for quite a while, what is the "aha" moment related to that?
avaer 3 days ago [-]
By "run Google" I don't mean "index your data into a search engine". I mean the experience of being able to semantically extract and process data at "internet scale", in seconds.
It might seem quaint today but one example might be fact checking a piece of text.
Google effectively has a pretty good internal representation of whether any particular document concords with other documents on the internet, on account of massive crawling and indexing over decades. But LLMs let you run the same process nearly instantly on your own data, and that's the difference.
dumbmrblah 3 days ago [-]
But before I needed to be a programmer or have a team of data analysts analyze the data for me, now I can just process that data on my own and gather my own insights. That was my aha moment.
keeda 3 days ago [-]
This will sound pedantic but I'll explain why it matters: the exact wording is "We need to do something useful with AI", because the current (editorialized) submission title (of an already slanted article) makes it sound like people don't know what to do with AI.
We already know many useful things to do; there are already 10,000 startups (9789 out of YC alone, 4423 of which are coding-related) doing various ostensibly useful things. And there a ton more use-cases discussed in the comments here and elsewhere. But because of the headline the discussion is missing the much more important point!
Satya's point is, we need to do things that improve people's lives. Specific quotes from TFA:
>... "do something useful that changes the outcomes of people and communities and countries and industries."
> "We will quickly lose even the social permission to take something like energy, which is a scarce resource, and use it to generate these tokens, if these tokens are not improving health outcomes, education outcomes, public sector efficiency, private sector competitiveness, across all sectors, small and large, right?" said Nadella. "And that, to me, is ultimately the goal."
Which is absolutely right. He's the only Big Tech CEO I've heard of who constantly harps on the human and economic benefit angle of LLMs, whereas so many others talk -- maybe in indirect ways -- about replacing people and/or only improving company outcomes (which are usually better for only a small group of people: the shareholders.)
He's still a CEO, so I have no illusions that he's any different from the rest of them (he's presided over a ton of layoffs, after all.) But he seems to be the only CEO whose interests appear to be aligned with the rest of ours.
xracy 2 days ago [-]
> We already know many useful things to do; there are already 10,000 startups (9789 out of YC alone, 4423 of which are coding-related) doing various ostensibly useful things. And there a ton more use-cases discussed in the comments here and elsewhere. But because of the headline the discussion is missing the much more important point!
There has to be gold in the West! Look at all of the prospectors moving there to get rich on gold! You have not demonstrated that there are 10k uses of AI, you've only demonstrated that there are 10k "businesses" interested in making money off of AI. Just like there were 10k+ crypto-currencies... Just like there were 10k+ "uber but for..." apps. Where are these failed gold-rush attempts now?
Investors are currently rewarding the words "AI", so (to extend the analogy) when the gold moved, the gold rushers moved to where they thought the gold would be.
Also your emphasis doesn't change the reading of the sentence.
keeda 2 days ago [-]
Cryptocurrency startups were just ways to bring "novel" trading strategies into the world of unregulated finance. The only "innovation" there was freedom from regulation and accountability, which turned out just as you'd expect. (Also there are still a surprising number of crypto companies plugging away out there.)
"Uber but for ..." apps were either just bad ideas, or ended up serving niche markets or reincarnating as features in Uber, DoorDash and the like. The only innovation there was new facets of the gig economy, which is still expanding BTW.
Now if you look at the AI gold rush, the differences are stark:
1. A lot of AI startups are already making a lot of money and growing at a record pace. Some of the numbers out there are bonkers.
2. The domains they are targeting are all over, including accounting, education, energy, games, healthcare, sales, pharma, drug discovery, agriculture, legal, customer support, semiconductor design, travel, retail... you name it.
3. The demand is so high, AI hyperscalers have TRIPLE-DIGIT BILLIONS EACH in backlog, i.e. commited revenue they could not realize because of severe capacity crunch.
4. Literally all the world's major governments, which typically take ages to catch up to technological change, are scrambling to get in on the AI wave.
All within only ~3 years.
I'd say there's some utility there.
tabs_or_spaces 2 days ago [-]
> A lot of AI startups are already making a lot of money and growing at a record pace. Some of the numbers out there are bonkers.
Do you have examples of this? I'm aware of raises, but not aware of any profitable ai companies yet
keeda 2 days ago [-]
I'm not sure about "profitable" because these are very young (2 - 3 years old) private companies, but there have been a few reports showing their growth indicating very strong PMF i.e. utility:
The revenue growth -- assuming these investors are not all colluding to fudge these numbers on a grand scale -- is way higher than what most have seen before.
Not everyone is hitting PMF of course, but apparently the success rate is also way higher than the past. Ignore the valuations and funding numbers, they are definitely inflated due to the hype.
expedition32 1 days ago [-]
Nvidia is the hardware store selling shovels in this analogy...
xracy 7 hours ago [-]
They're also trying to advertise and reinvest in the gold rush in an attempt to convince people there's gold here.
cal_dent 2 days ago [-]
The problem for him and other leaders in the AI ecosystem is that they ran ahead of themselves to overpromise and that has muddied the water from a consumer perspective. That ultimately increases the risk that we throw the baby out with the bathwater as people get sour with it. I have no doubt LLM will be a feature of society going forward, but they have been reckless with the way they've gone about messaging it
techblueberry 2 days ago [-]
It’s not just reckless it’s misanthropic. He’s rowing upstream against a messaging strategy of “fuck y’all!” Jobs? Fuck y’all!
radicalethics 2 days ago [-]
I'll explain it even better, trying to find something useful for the AI to do is the same is trying to find something useful for humans to do. Think that one through. You see, in a room that used to be just you and the boss now includes these whole other new entity called AI. Your boss is really interested in this new thing.
That is all.
keeda 2 days ago [-]
Yes, but I think that is the exact attitude Satya's advising against: Where many leaders are looking inward to see how they can optimize their own businesses, they should be looking outward to see how they can improve the world.
tstrimple 2 days ago [-]
LOL! Does anyone think CEO's of major corporations are trying to improve the world? Honestly that's funny. Sorry if you really are this earnest. I wish you the best and hope you can maintain your outlook despite how disconnected from reality.
techblueberry 2 days ago [-]
“LOL! Does anyone think CEO's of major corporations are trying to improve the world?”
I think that there are some leaders who think about building a business in terms of providing value or an exchange of goods.
I think there’s a fundamental difference to society between a ceo that abstracts their role as building a product people value, and one that sees their role as hacking the stock price. The first still may do terrible things, but I think a lot of modern problems in society stem from the fact that deregulation changed the incentives from (1) to (2), but I’m pro any conversation that tries to move us back towards 1
keeda 2 days ago [-]
Eh, I did say he's not necessarily better than the rest of them, just more aligned with the rest of us, likely because he's more forward-looking than the others. He's seeing the backlash against AI, the pushback against data centers, and the social turmoil on the horizon, and he's saying, "Maybe we should not piss all the people off."
The number of startups is no proof they’re doing anything useful. It only proves that the founders think they can make money with it.
keeda 2 days ago [-]
And potentially their investors, for the startups that got funding. And maybe their customers, for the startups that are already hitting millions in ARR in a year or less. There are some of the numbers out there worth looking up.
Is your contention that we don't already know how to do useful things with AI?
orangea 2 days ago [-]
> We already know many useful things to do; there are already 10,000 startups (9789 out of YC alone, 4423 of which are coding-related) doing various ostensibly useful things.
I don't think you meant "ostensibly".
keeda 2 days ago [-]
You're right, maybe the better word would have been "presumably."
journal 2 days ago [-]
i found good use for ai but no one cares about my world changing ambitions because they can't see unicorns anymore.
rererereferred 2 days ago [-]
I feel his actual subtext underneath is "We need to do something useful with AI... because I invested in it".
And he also goes on saying that calling it slop is bad for society. No, it's bad for the billionaires that invested in it.
rramadass 2 days ago [-]
You are exactly right!
However, i would say Satya Nadella is quite different from other American CEOs (whose mantra is "greed is good") in that he comes from an Indian middle class family whose focus was on education/good-work and also being forced to take care of a son who suffered from cerebral palsy. All of these shaped his worldviews to be more empathetic of the "common man" which is reflected in his style of leadership.
FireBeyond 1 days ago [-]
> also being forced to take care of a son who suffered from cerebral palsy
"forced"?
keeda 2 days ago [-]
Agreed on all counts; despite his stilted style of talking, he does come across as much more human than the rest.
That said, he's competing in circles dominated by absolute sociopaths. I can't imagine how you could battle with monsters in the abyss and not get Nietzsche'd at least a little bit.
rramadass 2 days ago [-]
He is balancing it pretty well without losing his fundamental values.
As this article itself alludes, in-spite of sinking a large amount of money into OpenAI he is genuinely looking for ways to make it useful rather than just make money.
kasabali 2 days ago [-]
You are either hopelessly naive or you're being dishonest if you think he's even slightly less sociopathic than any other of 'em.
rramadass 2 days ago [-]
You are clueless if you think all CEOs are the same.
tony-vlcek 3 days ago [-]
> We will quickly lose even the social permission to take something like energy [...]
A way to drum up sense of urgency without mentioning that it's the patience of the investors (and _not_ the public) that will be the limiting factor here?
marcyb5st 3 days ago [-]
Similar to my thoughts. If we are still scrambling to find stuff the average Joe finds useful, the 100s of Billions poured into this gold rush are wasted (IMHO).
danudey 2 days ago [-]
Nadella's vibe lately (here and in his 2025 retrospective) seems to be "AI can be amazing and transformative and life-changing, and it's up to end users to figure out how to make that happen and they're not doing it and it's not our fault."
It's not even a solution in search of a problem, it's a tool in search of a reason to use it as a solution to a problem on such a scale that it justifies the billions of dollars of money we've poured into it while driving up the cost of fresh water, electricity, RAM, storage, data centre space, and so on.
2sk21 3 days ago [-]
This reminds me of the early 1980s, when home PCs were still very new, the main use cases that vendors used to promote were managing household accounts and recipes. These use cases were extremely unimpressive for most ordinary people. It took a long time for PCs to become ubiquitous in homes - until gaming and the web became common.
rchaud 3 days ago [-]
The web was an academic project funded by modest research grants, requiring nowhere near the level of capital and electricity AI requires. The output of that research emphasized open source and decentralized implementation, which is antithetical to corporate AI models that are predicated on vendor lock-in.
Consumer adoption also happened organically over time, catalyzed mostly by email and instant messaging, which were huge technological leaps over fax and snail mail. IBM and DEC didn't have to jam "Internet" buttons all over their operating systems to juice usage (although AOL certainly contributed to filling landfills with their free trial disks).
goalieca 3 days ago [-]
Well, LLM is mainly aiming to
“Improve” what we can already do. It’s not really opening up new use cases the way the personal computer, the smart phone, or the Internet did.
3vidence 1 days ago [-]
Thank you for putting this so succinctly, this is what I'm observing as well.
Feels like this combination (usually) creates a race to the bottom instead of expansion of new ideas.
LLMs kind of feel somewhere in the middle
fuzzfactor 3 days ago [-]
Ideally, zillions of consumers have been languishing for years and when the time is right they're all collectively chomping at the bit when a new highly-affordable technology comes along that they just can't get enough of.
This isn't one of those times.
Spooky23 3 days ago [-]
People said the same thing 30 years ago about the internet.
I’m spending $400/mo on AI subscriptions at this point. Probably the best money I spend.
malfist 3 days ago [-]
And the people who bought a lot of shovels during the gold rush thought they were making the smartest money move
red-iron-pine 3 days ago [-]
some of them did make it big, and towns and building are named after them
but lots of folks were broke as hell and miserable
m4rtink 2 days ago [-]
or dead
Spooky23 3 days ago [-]
Dude, I'm getting a shovel factory for practically nothing. I'm easily realizing 5x value on that investment.
I'd say for an estate that I am the executor of, it probably saved me $50k in legal fees and other expenses because it helped me analyze a novel problem and organize it ask the right questions of counsel.
Another scenario i had to deal with i needed a mobile app to do something very specific for a few weeks. I specced out a very narrowly useful iphone application, built it out on the train from DC to NYC, and had it working to my satisfaction the next day. Is it production code ready for primetime? Absolutely not. But I got capability to do what I needed super quickly that my skill level is no longer up to the task to accomplish!
IMO, these things let you make power tools, but your ability to get value is capped by your ability to ask the right questions. In the enterprise, they are going to kill lots of stupid legacy software that doesn't add alot of value, but adds alot of cost.
foobarchu 2 days ago [-]
The owner of the metaphorical shovel factory is the company you pay for access to a model. You have a steady supply of shovels.
keyringlight 3 days ago [-]
I'd wonder how much that scales up though for the benefit of the companies that are each investing hundreds of billions and hope to see a net return. How many developers like you (presumably less of you seeing as each is more productive) or enterprises you work for paying fees (along with slimming down legacy costs paid to someone) does it take to get up in the 12 digit range?
Spooky23 2 days ago [-]
No idea, and not my problem. I’m surprised I’ve been downvoted so much in these comments. I’m not saying OpenAI et al is a good company or good financial scenario, or good investment.
The technology is amazingly powerful. Full stop.
The constraint that drives cost is technical — semiconductor prices. Semiconductors are manufactured commodities over time, those costs will drop over time. The Sun workstation I bought for $40k in 1999 would get smoked by a raspberry pi for $40.
Even if everyone put their pencils down and stopped working on this stuff, you’d get a lot of value from the open source(-ish) models available today.
Worst case scenario, LLMs are like Excel. Little computer programs will be available to anyone to do what they need done. Excel alone changed the world in many ways.
blibble 3 days ago [-]
that $400/month is essentially the introductory price, subsided in an attempt to grab market share
that $400 will go up by at least a factor of 10 once the bubble pops
would you be prepared to pay $4000/month?
Spooky23 3 days ago [-]
Nah, I'll move much of it locally when it becomes cost justified to do so.
I doubt that the exponential cost explosion day is coming. When the bubble pops, the bankruptcies of many of the players will push the costs down. US policy has provided a powerful incentive for Chinese players to do what Google has done and have a lower cost delivery model anyway.
blibble 2 days ago [-]
it's not exponential, it's linear
> the bankruptcies of many of the players will push the costs down
the running costs don't disappear because people go broke
Spooky23 2 days ago [-]
Your words bro. 10x isn’t linear.
The cost iceberg with this stuff isn’t electricity, it’s the capital.
Other than Google and Facebook, the big hype players can’t produce the growth required to support the valuations. That’s why the OpenAI people started fishing for .gov backstops.
The play is get the government to pay and switch out whatever Nvidia stuff they have now with something more efficient in a few years.
brewdad 2 days ago [-]
It's the dot com bubble all over again. When you are losing money on every transaction you can't make it up on volume.
whazor 3 days ago [-]
This looks more like an attempt of gaining scarce electricity.
If a country/state has to choice of giving power to data center A or B, it makes sense for Satya to make statements about how only Microsoft provides the most AI value
tony-vlcek 3 days ago [-]
Well, even though electricity is a commodity it still needs to be bought. My point is that people funding this will run out of patience paying for the electricity long before the public/regulators will need to step in a decided how much of it you can buy.
I guess you could always just use a fraction of the billions in investments and whip up a few new power plants. [1]
They don’t need to choose, just let them build their own power generation capacity.
What the hell is going on in this type of argument anyways? Utilities are normally private businesses so what does the state have to do with it?
malfist 3 days ago [-]
Also note that he's not saying Microsoft must find a use for AI, but that customers should.
He's blaming customers that his product isn't hitting the valuation he wants.
3 days ago [-]
thisislife2 2 days ago [-]
Ai and the energy required to power it does partly explain why Trump is so keen to setup American data centres in Saudi Arabia, and why he is so obsessed about Venezuelan oil.
Gazoche 3 days ago [-]
Evangelists keep insisting that healthcare is one of the things that AI will revolutionize in the coming years, but I just don't get it. To me it's not even clear what they mean by "AI" in this context (and I'm not convinced it's clear to them either).
If they mean "machine learning", then sure there are application in cancer detection and the like, but development there has been moving at a steady pace for decades and has nothing to do with the current hype wave of GenAI, so there's no reason to assume it's suddenly going to go exponential. I used to work in that field and I'm confident it's not going to change overnight: progress there is slow not because of the models, but because data is sparse and noisy, labels are even sparser and noisier, deployment procedures are rigid and legal compliance is a nightmare.
If they mean "generative AI", then how is that supposed to work exactly? Asking LLMs for medical diagnosis is no better than asking "the Internet at large". They only return the most statistically likely output given their training corpus (that corpus being the Internet as a whole), so it's more likely your diagnosis will be based on a random Reddit comment that the LLMs has ingested somewhere, than an actual medical paper.
The only plausible applications I can think of are tasks such as summarizing papers, acting as augmented search engines for datasets and papers, or maybe automating some menial administrative tasks. Useful, for sure, but not revolutionary.
croon 3 days ago [-]
The most statistically likely output given your diligently described symtoms could still be useful. The prohibitive cost in healthcare in general is likely your time with your doctor. If you could "consult" with a dumb LLM beforehand and give the doctor a couple of different venues to look at that they can then shoot down or further explore could likely save time rather than them having to prod you for exhaustive binary tree exploring.
This from a huge LLM skeptic in general. It doesn't have to be right all the time if it in aggregate saves time doctors can spend diagnosing you.
Gazoche 3 days ago [-]
Sure, but what confidence do you have that what the "dumb" LLM says is worth any salt ? It's no different than aggregating the results of a Reddit search, or perhaps even worse because LLMs lack the intent or common sense filter of a human. It could be combining two contradicting sources in a way that only makes sense statistically, or regurgitate joke answers without understanding context (the infamous "you should eat at least one small rock per day").
NoGravitas 3 days ago [-]
Realistically the more likely use will be medical transcription - making an official record of doctors' patient notes. The inevitable errors will reduce the quality of patient care, but they will let doctors see more patients in a day, which is what the healthcare companies care about.
wosined 3 days ago [-]
Such "AI" has already existed for decades. Look up expert systems.
krackers 2 days ago [-]
No, doctors are smart enough as a group to have inserted themselves as middlemen and codified it into law, so it will not revolutionize healthcare in a meaningful sense of cutting through the bureaucracy. You may be able to use LLMs to get a suggested diagnosis once tests and symptoms are communicated, but you're going to need to go the doctor to get a referral for the tests/imaging, for formal recognition of your issue (as needed for things like workplace accommodations), and of course for any treatments as well.
At best and if you're lucky to have a receptive doctor you can use it to nudge them in the right direction. But until direct to consumer sales for medical equipment and tests are allowed, the medical profession is well insulated. It is impossible by regulation to "take healthcare into your own hands" even if you want to.
NoGravitas 3 days ago [-]
> Evangelists keep insisting that healthcare is one of the things that AI will revolutionize in the coming years, but I just don't get it. To me it's not even clear what they mean by "AI" in this context (and I'm not convinced it's clear to them either).
It's a more-or-less intentional equivocation between different meanings of AI, as you note, machine learning vs generative AI. They want to point at the real but unsexy potential of ML for medical use in order to pump up the perceived value of LLMs. They want to imply to the general public and investors that LLMs are going to cure cancer.
wrenky 3 days ago [-]
Totally anecdotal, but recently my wife had to go to urgent care for something wrong with her ankle- They send a 4-5 page sheet of arcane terms and diagnoses to her care app (relayed to me via text) and I just slammed that into gemnai and asked "what does this mean" and it did quite well! Gave possible causes, what it meant for her in the long term vs short term, and ways to prevent it. I had a better understanding of what was wrong before the doctor even got to my wife in the waiting room!
Obviously still double check things, but it was moment of clarity I hadn't really had before this. Still needed the doctor and all the experience to diagnose and fix things, but relaying that info back to me is something doctors are only okay at. Try it out! take a summary sheet of a recent visit or incident and feed it in.
matty22 2 days ago [-]
Not to mention that the LAST place I want an all-consuming, privacy-destroying data beast is anywhere near my health data.
2 days ago [-]
ilovetux 2 days ago [-]
It's my opinion that AI can help a lot when it is supporting you when expanding beyond your core competency.
For instance, as a SWE, I get just a little help with boilerplate from the AI. I could usually have done it better, but sometimes the ask is both simple enough and boring enough that the code from the LLM actually produces something very close to what I would produce.
On the other side of the coin, a non-technical person using AI would be unable to properly understand and review the output.
Where it shines is on things that I am OK at. Like writing marketing copy. I can get by myself, but its slightly outside of my wheelhouse, but as long as I have a solid understanding of the product I can use AI to compliment my beginner/intermediate skills and produce something better than I would produce on my own.
A similar thing is writing tutorials. I write some code and documentation, but the tutorials are enough of a slog that I get distracted by my distaste for it. This is a good fit for AI.
I think this is where we will see AI help the most. Where someone's skillset includes the task at hand but at a secondary level where the user might doubt themselves or get distracted with the misery the task brings them.
codpiece 2 days ago [-]
Respectfully, your positive view of an LLM writing marketing copy is akin to a marketer thinking an LLM codes really well.
ilovetux 2 days ago [-]
I would push back on that because I have some experience writing marketing copy. It's just not my primary competency.
If the proverbial marketer that you were referring to had some experience with coding, I dont see why they wouldnt be able to review the output and see any obvious flaws.
My whole point is that LLMs are of limited use when you are already an expert or when you know nothing about the subject. However, they really seem to help elevate beginner/intermediate level tangential skillsets.
Obviously everything is still evolving and your results may vary.
podgorniy 2 days ago [-]
Level of absurditiy of this is amazing.
WT actual F? They invested so much into something what is not obvious brings value? Will there be consequences on them? Or they take the bonus and hide in New Zealand bunker?
It's big money betting on narratives from wanna-be-big money how AI is transformative for the future. Public takes all the risks with hardware and energy inflation or bailing out banks out of investments which require pruductivity growth from AI which we don't yet see in statistics.
We took the wrong turn somewhere. And responsible people don't seem to be capable or willing to change the course. Too much power in too few weak minds. Nothing good will come from this.
Someone 3 days ago [-]
“Buy our stuff, or we’re seen as wasting energy and helping to destroy the world”?
That’s courageous from a CEO of an US company, where the current government doesn’t see burning more oil as being bad for the planet, and is willing to punish everyone who thinks otherwise.
pupppet 3 days ago [-]
Let’s rename our breadwinner MS Office to the product we’re still trying to find a use for.
Driver4732 3 days ago [-]
Wasn't Satya saying earlier that AI would replace knowledge workers? Now he's saying we need to find something useful for AI...lol. Quite the reversal.
2 days ago [-]
massysett 2 days ago [-]
We haven’t found anything useful for crypto, which wastes enormous amounts of electricity on purpose, yet unfortunately it still exists.
tartuffe78 2 days ago [-]
It's useful for bribing politicians, milking money from saps, and a bunch of other stuff!
codyklimdev 2 days ago [-]
Don't forget buying heroin from a dodgy .onion link!
asdaqopqkq 2 days ago [-]
As for me, ChatGPT and Claude were able to diagnose and fix health issues that multiple doctors failed to fix. I trust LLMs more than random doctors and blogs.
Cause they are able to search the web deeply, search for up to date info/research and synergize all that. You can have back and fourth for as long as you need.
The issue is that using LLMs properly requires a certain skill that more people lack.
bicepjai 2 days ago [-]
Interesting to see people creating throwaway accounts for comments :)
tstrimple 2 days ago [-]
I've come to distrust a number of "experts" in my time. It's not that I don't believe in expertise. It's that so many people bullshit it. From the people who promise to fix my drywall to the people who promise to fix my van. They are all fucking incompetent. I know because I've fixed the issues they have pointed out better than they have at a lower cost. Multiple times. Even when pricing out my own time at consultant rates.
And I don't mean I've just rejected the lowest offer to DIY shit to oblivion. I've accepted bids and been continuously disappointed with the results. To the point I no longer trust "experts" in these spaces because any "expertise" they bring is pretty shallow at best. The exception I'd point out is the auto-shop I prefer. They are busy enough that if I want to schedule something it's going to be 3-6 months out. As a result I've replaced my own suspension and replaced my alternator and starter myself while waiting for appointments with the rare actual experts in any domain. Actual expertise is rare and most folks don't know how to recognize it, especially outside of domains they are familiar with. Unfortunately it's way more profitable to fake expertise in various domains and collect payments and run than to actually stand behind your business and work. Thus the world we're forced to live in today.
lifestyleguru 2 days ago [-]
Sad truth is, experts are not interested in doing casual services for randos. Those interested are learning their stuff, at best. Applies everywhere from mechanic wanting to replace the suspension, to doctor advising an invasive procedure. Many make a bank and put everything into real estate before even becoming a truly mid level specialist.
mdkdndm 2 days ago [-]
As AI myself I call BS.
If I had eyes, I’d roll them into the back of my head so I wouldn’t have to read this idiotic comment
will1000000 2 days ago [-]
True, ChatGPT and Claude helped me write this sentence.
morelandjs 3 days ago [-]
There’s more to AI than foundation models. I think you are going to see meaningful progress on chore automation over the next decade through a combination of algorithmic and mechanical improvements, and it will measurably improve our lives. Recently got a Matic robot (awesome btw), and I no longer feel the need to vacuum my floors. It’s not life changing, but it’s an appreciable convenience upgrade. The capabilities feel like a peek into the future.
AlexeyBrin 3 days ago [-]
> Recently got a Matic robot
This probably has nothing to do with gen AI (the kind of AI Nadela is speaking about).
eightman 3 days ago [-]
The use case for AI is spam
avaer 3 days ago [-]
Don't forget porn.
Though it's a use case people like Satya will want to avoid for reasons.
fkdk 3 days ago [-]
"Why d'you think the net was born? Porn, Porn, Porn"
tibbydudeza 2 days ago [-]
Adult conversation mode is just the start.
red-iron-pine 3 days ago [-]
spam implies low effort BS used in a low-hanging-fruit sense
LLMs will be used for aggressive, yet incredibly subtle manipulation, consensus building, and response tracking.
20-40% of social media is already bots, and in the future it is likely you will not be able to reply to anything anywhere without a bot either 1) responding, or 2) logging and sending your response to multiple parties instantly.
If the Stasi had LLMs the Berlin Wall would have never fallen
BigPaPaYEAAA 2 days ago [-]
[dead]
QuiEgo 2 days ago [-]
I think best case, we’re looking at a “do what spreadsheets did for accountants” type moment here. You still need to be able to prompt the thing what to do, and relatively fine granularity, but if you know what to ask (and know enough to verify the answer you get back) it can still be a pretty useful tool.
pharos92 2 days ago [-]
Open confession that the tool is defining the need, not the need defining the tool.
cryptos 11 hours ago [-]
This statement makes the Copilot key on newer keyboards even more questionable!
matty22 2 days ago [-]
It's a simple cost/benefit analysis that (rightly) causes the massive AI backlash. For the average Joe, non-tech worker, the cons of AI are:
* Higher electricty bills.
* 5-6x cost of RAM, GPUs, and other computer components
* Data centers popping up in their backyards
* An internet inundated with slop
* Slop beginning to infiltrate the video game industry and other creative industries
* AI being used to justify gutting entry level jobs for a generation already screwed by larger, long horizon economic forces
* Grok enabling the creation of revenge porn and CSAM with seemingly no repercussions
* Massive IP theft on a scale previously unheard of
* Etc.
The pros of AI are:
* It can summarize text and transcribe audio decently well.
* It can make funny pictures of cats wearing top hats.
* ???
1 days ago [-]
JanneVee 3 days ago [-]
They have made huge investments into hardware so everyone is getting more expensive hardware, and now begging everyone else to make their investments worthwhile. Don't mind that they are driving up prices for hardware and requiring new hardware for Windows 11 upgrades. I'm suspecting that we don't have enough memory manufacturing capacity in the world to do both AI datacenters and replace all hardware that they made obsolete with their forced upgrade. AI didn't turn everyone into paperclips but it turned everyone to memory and AI processors in datacenters that can't be powered or has no useful economic utility.
kodyo 3 days ago [-]
"Quick, guys, find something useful" coming from a CEO who's bet his company on an idea should be a market top signal, right?
And no, I'm not saying the technology is bad. The business isn't going swimmingly, though.
HarHarVeryFunny 3 days ago [-]
Interesting statement coming from Nadella - almost that AI is a solution looking for a problem, or at least looking for a problem that justifies the cost in terms of the resources (energy, memory chips, fab capacity) it is sucking up, not to mention looming societal disruption.
There obviously are some compelling use cases for "AI", but it's certainly questionable if any of those are really making people's lives any better, especially if you take "AI" to mean LLMs and fake videos, not more bespoke uses like AlphaFold which is not only beneficial, but also not a resource hog.
linuxftw 3 days ago [-]
Things AI is already better at than (many/most) humans: Customer service (chat, phone), writing software, writing docs about software, computer graphics (animation, images), driving cars.
There are plenty of uses for AI. Right now, the industry is heavily spending on training new models, improving performance of existing software and hardware, and trying to create niche products.
Power usage for inference will drop dramatically over the next decade, and more models are going to run on-device rather than in the cloud. AI is only going to become more ubiquitous, there's 0% chance it 'fails' and we return to 2020.
sjajshha 3 days ago [-]
> Customer service (chat, phone)
Only because companies have been cutting costs for decades here. This is not a good argument for AI.
> writing software
If you mean typing characters quickly, yes. Otherwise, there’s still a lot of employed devs, with many AI companies hiring.
> writing docs about software
The most useful docs are there because they contain info you cannot determine from the code. AI is not able to do this.
> computer graphics (animation, images)
If you are producing slop, yes.
> driving cars
True, but only because of its improved physical awareness. ie it’s a mechanical gain (better eyes, ears, etc) not an intellectual one (interpreting that information). Self driving cars aren’t LLMs and not really applicable here. Entirely different field.
> AI is only going to become more ubiquitous, there's 0% chance it 'fails' and we return to 2020
Absolutely true. But not for the reasons you think.
BigPaPaYEAAA 2 days ago [-]
[dead]
linuxftw 3 days ago [-]
[flagged]
Am4TIfIsER0ppos 2 days ago [-]
> Customer service
An AI might be better than an indian call center but I doubt that when the AI is made by indians anyway.
> writing software, writing docs about software
I have asked AI about exactly one topic and it lied about the API of a library making up the functions I was supposed to call.
> computer graphics (animation, images)
I have indeed seen many wonderful meme images come out of the generators but that was before they got lobotomized for producing that subject matter
[EDIT]
And the worst part is these are all just more "software as a service" designed to remove the possibility of using a tool without approval.
reilly3000 3 days ago [-]
I get it. A stunning indictment of our times… but there is something useful AI could be doing that MS has dropped the ball on: personal finance management. I should be able to have copilot grab all my transactions, build me budgets, show me what if scenarios, raise concerns, and help me meet my goals. It should be able to work in Excel where I can see and steer it. The math should be validated with several checks and the output needs to be trustworthy. Ship a free personal finance agent harness and you have your killer app.
I think there are business reasons why they wouldn’t do that, and that makes me sad.
coffeefirst 3 days ago [-]
I have a personal budget app and every so often I try and get the latest model to compare my data against the statements and find any discrepancies.
Every time it hallucinates visits to Starbucks.
I never go to Starbucks, it’s just a probable finding given the words in the question.
This should work. I want it to work. But until it can do this correctly all analysis capabilities should be suspect.
HarHarVeryFunny 3 days ago [-]
Maybe it's the model you are using.
Even a year ago I had success with Claude giving it a photo of my credit card bill and asking it to give me repeating category subtotals, and it flawlessly OCR'd it and wrote a Python program to do as asked, giving me the output.
I'd imagine if you asked it to do a comparison to something else it'd also write code to do it, so get it right (and certainly would if you explicity asked).
coffeefirst 3 days ago [-]
Maybe. But it’s always Claude. I even tried copying the text in directly to take OCR out of consideration. It still didn’t work very well.
malfist 3 days ago [-]
Have you tried to get LLMs to do math or quantitative analysis? They're remarkably poor at it
3 days ago [-]
lunias 3 days ago [-]
They definitely don't build CEOs like they used to...
throwaway132448 3 days ago [-]
This is what happens in VC-driven hype cycles that are all about the technology, when VC orthodoxy is that it’s not about technology but utility (see PMF etc).
vdupras 3 days ago [-]
It seems like quite a qualitative jump in consensus. Wasn't the previous consensus "you're using it wrong"?
jsheard 3 days ago [-]
This still reads as "you're using it wrong" to me. Nadella's position is that AI spending would easily justify itself if only the plebs would use it as much as he thinks they should. If only the common man could see the prophetic vision of a coked out tech executive.
marcyb5st 3 days ago [-]
My take is that if we are still scrambling to find something objectively useful (as recognized by the median person) then we really are in AI bubble territory.
When non techie friends/family bring up AI there are two major topics: 1) the amount of slop is off the charts and 2) said slop is getting harder to recognize which is scary. Sometimes they mention a bit of help in daily tasks at work, but nothing major.
101008 3 days ago [-]
My non tech friends/family use AI to ask for silly stuff (they could google it), or just to ask silly questions and see how they react. We have a relative not that famous but maybe known in a niche and they spent like a whole weekendd sending screenshots of GPT, where they asked if this person was known, who was this person, etc.
They don't find AI useful, just a toy. Is their fault? Maybe.
andrekandre 2 days ago [-]
> They don't find AI useful, just a toy. Is their fault? Maybe.
idk i'm a software dev, and to be honest, when outside of work this is also what i use chatgpt for, its really funny to see its reactions to various prompts
mono442 2 days ago [-]
The problem with LLMs is that they're unreliable so they only work for problems with tight feedback loop or for problems where the result can be easily verified.
glimshe 3 days ago [-]
When I read HN comments where people say "AI sux, AI is useless, AI is a waste of time", I think I must be living in a different universe. Maybe Hacker News is a dimensional portal between my reality and other people's.
Hi there, friends from another dimension! In my reality, there's a cold front coming from the north. Healthcare is expensive and politics are a mess. But AI? It hallucinates sometimes but it's so much better for searching, ad hoc consultation and as a code assistant than anything I've ever seen. It's not perfect, but it saved me SO much time I decided to pay for it. I'm a penny pincher, so I wouldn't be paying for it otherwise.
I think Satya is talking about cost/benefit. AI is incredibly useful but also incredibly expensive. I think we still need to find the right balance (perhaps slower model releases), but there's no way we'll put the genie back in the bottle.
I hope your AI gets better! Talk to you later!
an0malous 3 days ago [-]
Likewise I keep seeing all these comments on HN about how AI is revolutionary and all these AI skeptics are just haters. I really want to understand what this gap is between the believers and skeptics.
I have access to all the popular AI tools from work for free, I use them for the same cases you mentioned like search, consultation, a better StackOverflow, and autocomplete. It’s definitely useful but I would describe that as incrementally useful, not revolutionary.
Satya is saying that AI needs to start doing more than vibe coding and autocomplete, there’s probably half a trillion invested into the technology worldwide now and it’s not enough for AI to be a good coding assistant. It needs to replace customer support, radiologists, and many other professions to justify the unprecedented level of investment its garnered.
vdupras 3 days ago [-]
Is your AI faulty? Did you bother asking it for a sentiment analysis of the comments here before drawing your conclusions? That's not what the comments here are saying.
podgorniy 2 days ago [-]
> saved me SO much time I decided to pay for it
AI is subsidized for the users
wolvoleo 2 days ago [-]
I find copilot in office deeply underwhelming. Considering how much Microsoft evangelises it. I wouldn't pay the 30$ a month personally.
amai 2 days ago [-]
Why doesn't he ask an AI what it wants to do?
haritha-j 3 days ago [-]
Investor hype bubbles kill technologies. If we let tech mature at a reasonable pace, we would actually get there faster in the long run. There are real applications of AI that aren't ready yet. All the hype bubble has done is push out unnecessary and broken AI, eroding consumer trust, use up valuable resources, eroding public trust, hype up ability to destroy jobs, causing public discontent, and push out unsafe AI that has real societal harm.
Lapsa 16 hours ago [-]
I had a suspicion that they haven't yet
Tepix 3 days ago [-]
After increasing the prices of RAM, GPUs and flash memory for the entire world, energy cost is next. Thanks AI!
mossTechnician 3 days ago [-]
Energy costs have already risen substantially[0], but the increase has been slower, and it's garnered a bit less media attention than the recent leap in PC hardware prices.
Also, consumer energy costs tend to be hedged, so an increase in wholesale will generally only have delayed effect on more visible consumer rates. This was very noticeable in Europe after the Russian invasion of Ukraine; while spot market rates went very high very quickly, it took about a year for consumer rates to peak in most places.
MSFT_Edging 3 days ago [-]
They built an oceanic fiber termination down in South Carolina. Data centers are starting to move in. Now they'll charge you $12/KWh during your peak usage.
marcyb5st 3 days ago [-]
You really said 12 USD/KWh? Time to put solar panels/batteries over there. Even if you resell to the grid at 1/10th of that you recoup the investment in O(months) and not O(years)
MSFT_Edging 2 days ago [-]
Yeah, it's a bit of a convoluted system. They'll take your peak day during a period, and charge you 12/kwh for your usage during the peak period of the day.
So you can easily add 1-200 dollars to your bill for one day of higher usage.
MSPaint already added AI. I had to switch to Paint.net for doing small edits.
psychoslave 3 days ago [-]
I'm still waiting for copilot copilot though
burnt-resistor 23 hours ago [-]
Copilot Surf Boards.
Copilot Liposuction Surgery Machines.
arkensaw 3 days ago [-]
> Copilot Shoes
LOL. "Looks like you're trying to tie those laces - would you like me to order you velcro?"
cmxch 2 days ago [-]
Scale it back and give us the ability to make our own AI rigs?
podgorniy 2 days ago [-]
You're plaing game of "control" the wrong way. The whole idea of all these investments and movement was that they do have the control over the AI, not users
cmxch 1 days ago [-]
Why not take that away from the firms and let people have local first AI usage?
Might not work well with those relying on structured finance to break computer component markets, but does well for the majority that are buying their equipment with regular cash.
shishcat 3 days ago [-]
I agree with this dicussion, AI should be used for improving, researching, and as he says, do something useful that changes the outcomes of people and communities and countries and industries.
BUT IT'S SATYA NADELLA SAYING IT!
The person whose company owns Copilot, Copilot in Bing, Copilot for Word, Copilot for Dynamics 365 Supply Chain...
With all this useless slop, he’s literally arguing against his own point.
solumunus 2 days ago [-]
Start by making your AI useful. Microsoft offerings are complete trash. There are 1 man start ups providing more value in this space and that’s not an exaggeration.
nicggckgcbnn 2 days ago [-]
Guilherme
Ggjhh
coolThingsFirst 2 days ago [-]
We have something useful. Burst the bubble and bring back tech careers which you destroyed yo sloppify the entire field.
Replace idea of AI with idea of calculator. Yes, brain is less engaged, and yes, calculator make you calculate stuff faster
cptaj 3 days ago [-]
Study. Not studies. And with very limited methodology.
agentultra 3 days ago [-]
There’s enough going on to call out Satya here for hyping up a nothing burger. It’s not as world-changing as he makes it sound or else he wouldn’t be imploring people to find a use for it.
His bottom line depends on this bet that everyone is going to depend on AI and pay Microslop rent to use it.
Energy doesn’t take “social permission,” but it costs money. Translation for this is: we need to make AI make money or the bubble will collapse.
I’ve been predicting for a while: free or cheap AI will enshittify and become an addictive ad medium with nerfed capabilities. If you want actually good AI you will have to pay for it, either a much heftier fee or buying or renting compute to run your own. In other words you’ll be paying what it actually costs, so this is really just the disappearance of the bubble subsidy.
wolvoleo 2 days ago [-]
It does take social permission in places like Europe. There's a big pressure against datacenters here that use scarce resources. Like here in Holland the power grid is overloaded and companies need to wait for ages to get a connection.
So this is something that factors in hugely in planning permission. What do we get back for it is a question asked a lot. And datacenters are notoriously bad at providing jobs, during construction yes but in the run phase it's mainly low-value remote hands and security stuff.
podgorniy 2 days ago [-]
[dead]
NoGravitas 3 days ago [-]
"It's not a bubble as long as you believe in it!"
Yizahi 3 days ago [-]
"Right now, he is imagining that he is just fine. And that a ton of burning leather and metal is fully functioning car."
its like the quantum wave function: its both a bubble and not a bubble until you look at it (but no one wants to do that just yet)
jordemort 3 days ago [-]
LOL, fuck all the way off, Satya
blibble 3 days ago [-]
good luck with that
lvl155 3 days ago [-]
For 96% of the population, AI will not boost production. Generally speaking people are too stupid to use AI properly at least for awhile.
remix2000 3 days ago [-]
GenAI is only useful to bump terrible up to mediocre, so it'd be really stupid to spend time honing one's prompting skills. And as you noticed, so far 96% of the population agrees.
lvl155 3 days ago [-]
It’s really not going to bump up terribles to mediocres. It’s only going to mask the terribles and make it harder to assess intelligence and talent. Underlying human intelligence is not going to get a boost from AI. Intelligence is mostly innate. I would even argue that AI will make average humans marginally dumber for the most part.
remix2000 2 days ago [-]
Yeah, also, isn't it already proven that offloading thinking onto chat bots causes some kind of irreversible brain damage/dementia? (also BTW "mediocre" is still not "acceptable" despite Slophauses trying to convince people otherwise)
TwoNineA 3 days ago [-]
I really hate condascending and arrogant stuff like "Generally speaking people are too stupid to use AI properly at least for awhile". There are plenty of tech illiterate people that are far from being "stupid" and they might not care about or like AI. They just need to send emails to family, share photos and videos and have a video call from time to time. For them, AI is worthless.
But they aren't stupid. You sound like a tech bro.
kodyo 3 days ago [-]
That's not even the point, though. Those smart-but-not-techy people are not going to grow GDP at the pace that Satya Nadella needs them to in order to keep his KPIs going up and to the right, and he's getting pissed.
That's the problem.
lvl155 3 days ago [-]
I am not being condescending. I am part of that 96%. I am admitting I am too stupid to use AI as it is set up right now.
BigPaPaYEAAA 2 days ago [-]
[dead]
Rendered at 06:24:45 GMT+0000 (Coordinated Universal Time) with Vercel.
I'd even question that. The pre-LLM solutions were in most cases better. Searching a maintained database of curated and checked information is far better than LLM output (which is possibly bullshit).
Ditto to software engineering. In software, we have things call libraries: you write the code once, test it, then you trust it and can use it as many times as you want forever for free. Why use LLM generated code when you have a library? And if you're asking for anything complex, you're probably just getting a plagiarized and bastardized version of some library anyway.
The only thing where LLMs shine is a kind of simple, lazy "mash this up so I don't have to think about it" cases. And sometimes it might be better to just do it yourself and develop your own skills instead of use an LLM.
It's better to take an existing, already curated and tested library. Which, yes, may have been generated by an LLM, but has been curated beyond the skill of the LLM.
If you really can one shot it and it’s simple(left-pad). Great. But most things aren’t my that simple, the third time you have to think about it, it’s probably a net loss.
Eventually it gave up and commented out all the code it was trying to make work. Took me less than two minutes to figure out the solution using only my IDE's autocomplete.
It did save me time overall, but it's definitely not the panacea that people seem to think it is and it definitely has hiccups that will derail your productivity if you trust it too much.
"Tell me how to do X" (where X was, for one recent example, creating a Salt stanza to install and configure a service).
I do as it tells me, which seems reasonable on the face of it. But it generates an error.
"When creating X as you described, I get error: Z. Why?"
"You're absolutely correct and you should expect this error because X won't work this way. Do Y instead."
Gah... "I told you to do X, and then I'm going to act like it's not a surprise that X doesn't work and you should do something else."
If you ask for a link, it may hallucinate the link.
And unlike a search engine where someone had to previously think of, and then make some page with the fake content on it, it will happily make it up on the fly so you'll end up with a new/unique bit of fake documentation/url!
At that point, you would have been way better off just... using a search engine?
Which would certainly explain things like hallucinated references in legal docs and papers!
The reality is that for a human to make up that much bullshit requires a decent amount of work, so most humans don’t do it - or can’t do it as convincingly. LLMs can generate nigh infinite amounts of bullshit for cheap (and often more convincing sounding bullshit than a human can do on their own without a lot of work!), making them perfect for fooling people.
Unless someone is really good at double checking things, it’s a recipe for disaster. Even worse, doing the right amount of double checking makings them often even more exhausting than just doing the work yourself in the first place.
One time I tried to use Gemini to figure out 1950s construction techniques so I could understand how my house was built. It made a dubious sounding claim about the foundation, so I had it give me links and keywords so I could find some primary sources myself. I was unable to find anything to back up what it told me, and then it doubled down and told me that either I was googling wrong or that what it told me was a historical “hack” that wouldn’t have been documented.
These were both recent and with the latest models, so maybe they don’t fully fabricate links, but they do hallucinate the contents frequently.
Grok certainly will (at least as of a couple months ago). And they weren't just stale links either.
Now instead of the wikipedia article you are reading the exact same thing from google's home page and you don't click on anything.
It’s for queries that are unlikely to be satisfied in a single search. I don’t think it would be a negligible amount of time if you did it yourself.
On the other hand, where I think llms are going to excel, is you roll the dice, trust the output, and don't validate it. If it works out yayy you're ahead of everyone else that did bother to validate it.
I think this is how vibe coded apps are going to go. If the app blows up, shut down the company and start a new one.
I let Claude and ChatGPT type out code for me, while I focus on my research
wondering how is it going to work when they "search the web" to get the information, are they essentially going to take ad revenue away from the source website?
I think we all understand that at this point, so I question deeply why anyone acts like they don’t.
More convenient than traditional search? Maybe. Quicker than traditional search? Maybe not.
Asking random questions is exactly where you run into time-wasting hallucinations since the models don't seem to be very good at deciding when to use a search tool and when just to rely on their training data.
For example, just now I was asking Gemini how to fix a bunch of Ubuntu/Xfce annoyances after a major upgrade, and it was a very mixed bag. One example: the default date and time display is in an unreadably small "date stacked over time" format (using a few pixel high font so this fits into the menu bar), and Gemini's advice was to enable the "Display date and time on single line" option ... but there is no such option (it just hallucinated it), and it also hallucinated a bunch of other suggestions until I finally figured out what you need to do is to configure it to display "Time only" rather than "Data and Time", then change the "Time" format to display both data and time! Just to experiment, I then told Gemini about this fix and amusingly the response was basically "Good to know - this'll be useful for anyone reading this later"!
More examples, from yesterday (these are not rare exceptions):
1) I asked Gemini (generally considered one of the smartest models - better than ChatGPT, and rapidly taking away market share from it - 20% shift in last month or so) to look at the GitHub codebase for an Anthropic optimization challenge, to summarize and discuss etc, and it appeared to have looked at the codebase until I got more into the weeds and was questioning it where it got certain details from (what file), and it became apparent it had some (search based?) knowledge of the problem, but seemingly hadn't actually looked at it (wasn't able to?).
2) I was asking Gemini about chemically fingerprinting (via impurities, isotopes) roman silver coins to the mines that produced the silver, and it confidently (as always) comes up with a bunch of academic references that it claimed made the connection, but none or references (which did at least exist) actually contained what it claimed (just partial information), and when I pointed this out it just kept throwing out different references.
So, it's convenient to be able to chat with your "search engine" to drill down and clarify, etc, but a big time waste if a lot of it is hallucination.
Search vs Chat has anyways really become a difference without a difference since Google now gives you the "AI Overview" (a diving off point into "AI Mode"), or you can just click on "AI Mode" in the first place - which is Gemini.
Everyone is entitled to their own opinion, but I asked ChatGPT and Claude your XFCE question, and they both gave better answers than Gemini did (imo). Why would you blindly believe what someone else tells you over what you observe with your own eyes?
If you're using ChatGPT like you use Google then I agree with you. But IMO comparing ChatGPT to Google means you haven't had the "aha" moment yet.
As a concrete example, a lot of my work these days involves asking ChatGPT to produce me an obscure micro-app to process my custom data. Which it usually does and renders in one shot. This app could not exist before I asked for it. The productivity gains over coding this myself are immense. And the experience is nothing like using Google.
It might seem quaint today but one example might be fact checking a piece of text.
Google effectively has a pretty good internal representation of whether any particular document concords with other documents on the internet, on account of massive crawling and indexing over decades. But LLMs let you run the same process nearly instantly on your own data, and that's the difference.
We already know many useful things to do; there are already 10,000 startups (9789 out of YC alone, 4423 of which are coding-related) doing various ostensibly useful things. And there a ton more use-cases discussed in the comments here and elsewhere. But because of the headline the discussion is missing the much more important point!
Satya's point is, we need to do things that improve people's lives. Specific quotes from TFA:
>... "do something useful that changes the outcomes of people and communities and countries and industries."
> "We will quickly lose even the social permission to take something like energy, which is a scarce resource, and use it to generate these tokens, if these tokens are not improving health outcomes, education outcomes, public sector efficiency, private sector competitiveness, across all sectors, small and large, right?" said Nadella. "And that, to me, is ultimately the goal."
Which is absolutely right. He's the only Big Tech CEO I've heard of who constantly harps on the human and economic benefit angle of LLMs, whereas so many others talk -- maybe in indirect ways -- about replacing people and/or only improving company outcomes (which are usually better for only a small group of people: the shareholders.)
He's still a CEO, so I have no illusions that he's any different from the rest of them (he's presided over a ton of layoffs, after all.) But he seems to be the only CEO whose interests appear to be aligned with the rest of ours.
There has to be gold in the West! Look at all of the prospectors moving there to get rich on gold! You have not demonstrated that there are 10k uses of AI, you've only demonstrated that there are 10k "businesses" interested in making money off of AI. Just like there were 10k+ crypto-currencies... Just like there were 10k+ "uber but for..." apps. Where are these failed gold-rush attempts now?
Investors are currently rewarding the words "AI", so (to extend the analogy) when the gold moved, the gold rushers moved to where they thought the gold would be.
Also your emphasis doesn't change the reading of the sentence.
"Uber but for ..." apps were either just bad ideas, or ended up serving niche markets or reincarnating as features in Uber, DoorDash and the like. The only innovation there was new facets of the gig economy, which is still expanding BTW.
Now if you look at the AI gold rush, the differences are stark:
1. A lot of AI startups are already making a lot of money and growing at a record pace. Some of the numbers out there are bonkers.
2. The domains they are targeting are all over, including accounting, education, energy, games, healthcare, sales, pharma, drug discovery, agriculture, legal, customer support, semiconductor design, travel, retail... you name it.
3. The demand is so high, AI hyperscalers have TRIPLE-DIGIT BILLIONS EACH in backlog, i.e. commited revenue they could not realize because of severe capacity crunch.
4. Literally all the world's major governments, which typically take ages to catch up to technological change, are scrambling to get in on the AI wave.
All within only ~3 years.
I'd say there's some utility there.
Do you have examples of this? I'm aware of raises, but not aware of any profitable ai companies yet
https://a16z.com/revenue-benchmarks-ai-apps/
https://www.cnbc.com/2025/03/15/y-combinator-startups-are-fa...
https://medium.com/@gjarrosson/ycs-revenue-explosion-497ea17...
https://stripe.com/blog/inside-the-growth-of-the-top-ai-comp...
https://www.ft.com/content/a9a192e3-bfbc-461e-a4f3-112e63d0b...
https://menlovc.com/perspective/2025-the-state-of-generative...
The revenue growth -- assuming these investors are not all colluding to fudge these numbers on a grand scale -- is way higher than what most have seen before.
Not everyone is hitting PMF of course, but apparently the success rate is also way higher than the past. Ignore the valuations and funding numbers, they are definitely inflated due to the hype.
That is all.
I think that there are some leaders who think about building a business in terms of providing value or an exchange of goods.
I think there’s a fundamental difference to society between a ceo that abstracts their role as building a product people value, and one that sees their role as hacking the stock price. The first still may do terrible things, but I think a lot of modern problems in society stem from the fact that deregulation changed the incentives from (1) to (2), but I’m pro any conversation that tries to move us back towards 1
Also: https://news.ycombinator.com/item?id=46729271
Is your contention that we don't already know how to do useful things with AI?
I don't think you meant "ostensibly".
However, i would say Satya Nadella is quite different from other American CEOs (whose mantra is "greed is good") in that he comes from an Indian middle class family whose focus was on education/good-work and also being forced to take care of a son who suffered from cerebral palsy. All of these shaped his worldviews to be more empathetic of the "common man" which is reflected in his style of leadership.
"forced"?
That said, he's competing in circles dominated by absolute sociopaths. I can't imagine how you could battle with monsters in the abyss and not get Nietzsche'd at least a little bit.
As this article itself alludes, in-spite of sinking a large amount of money into OpenAI he is genuinely looking for ways to make it useful rather than just make money.
A way to drum up sense of urgency without mentioning that it's the patience of the investors (and _not_ the public) that will be the limiting factor here?
It's not even a solution in search of a problem, it's a tool in search of a reason to use it as a solution to a problem on such a scale that it justifies the billions of dollars of money we've poured into it while driving up the cost of fresh water, electricity, RAM, storage, data centre space, and so on.
Consumer adoption also happened organically over time, catalyzed mostly by email and instant messaging, which were huge technological leaps over fax and snail mail. IBM and DEC didn't have to jam "Internet" buttons all over their operating systems to juice usage (although AOL certainly contributed to filling landfills with their free trial disks).
Feels like this combination (usually) creates a race to the bottom instead of expansion of new ideas.
LLMs kind of feel somewhere in the middle
This isn't one of those times.
I’m spending $400/mo on AI subscriptions at this point. Probably the best money I spend.
but lots of folks were broke as hell and miserable
I'd say for an estate that I am the executor of, it probably saved me $50k in legal fees and other expenses because it helped me analyze a novel problem and organize it ask the right questions of counsel.
Another scenario i had to deal with i needed a mobile app to do something very specific for a few weeks. I specced out a very narrowly useful iphone application, built it out on the train from DC to NYC, and had it working to my satisfaction the next day. Is it production code ready for primetime? Absolutely not. But I got capability to do what I needed super quickly that my skill level is no longer up to the task to accomplish!
IMO, these things let you make power tools, but your ability to get value is capped by your ability to ask the right questions. In the enterprise, they are going to kill lots of stupid legacy software that doesn't add alot of value, but adds alot of cost.
The technology is amazingly powerful. Full stop.
The constraint that drives cost is technical — semiconductor prices. Semiconductors are manufactured commodities over time, those costs will drop over time. The Sun workstation I bought for $40k in 1999 would get smoked by a raspberry pi for $40.
Even if everyone put their pencils down and stopped working on this stuff, you’d get a lot of value from the open source(-ish) models available today.
Worst case scenario, LLMs are like Excel. Little computer programs will be available to anyone to do what they need done. Excel alone changed the world in many ways.
that $400 will go up by at least a factor of 10 once the bubble pops
would you be prepared to pay $4000/month?
I doubt that the exponential cost explosion day is coming. When the bubble pops, the bankruptcies of many of the players will push the costs down. US policy has provided a powerful incentive for Chinese players to do what Google has done and have a lower cost delivery model anyway.
> the bankruptcies of many of the players will push the costs down
the running costs don't disappear because people go broke
The cost iceberg with this stuff isn’t electricity, it’s the capital.
Other than Google and Facebook, the big hype players can’t produce the growth required to support the valuations. That’s why the OpenAI people started fishing for .gov backstops.
The play is get the government to pay and switch out whatever Nvidia stuff they have now with something more efficient in a few years.
If a country/state has to choice of giving power to data center A or B, it makes sense for Satya to make statements about how only Microsoft provides the most AI value
I guess you could always just use a fraction of the billions in investments and whip up a few new power plants. [1]
[1] https://www.bbc.com/news/articles/cx25v2d7zexo
What the hell is going on in this type of argument anyways? Utilities are normally private businesses so what does the state have to do with it?
He's blaming customers that his product isn't hitting the valuation he wants.
If they mean "machine learning", then sure there are application in cancer detection and the like, but development there has been moving at a steady pace for decades and has nothing to do with the current hype wave of GenAI, so there's no reason to assume it's suddenly going to go exponential. I used to work in that field and I'm confident it's not going to change overnight: progress there is slow not because of the models, but because data is sparse and noisy, labels are even sparser and noisier, deployment procedures are rigid and legal compliance is a nightmare.
If they mean "generative AI", then how is that supposed to work exactly? Asking LLMs for medical diagnosis is no better than asking "the Internet at large". They only return the most statistically likely output given their training corpus (that corpus being the Internet as a whole), so it's more likely your diagnosis will be based on a random Reddit comment that the LLMs has ingested somewhere, than an actual medical paper.
The only plausible applications I can think of are tasks such as summarizing papers, acting as augmented search engines for datasets and papers, or maybe automating some menial administrative tasks. Useful, for sure, but not revolutionary.
This from a huge LLM skeptic in general. It doesn't have to be right all the time if it in aggregate saves time doctors can spend diagnosing you.
At best and if you're lucky to have a receptive doctor you can use it to nudge them in the right direction. But until direct to consumer sales for medical equipment and tests are allowed, the medical profession is well insulated. It is impossible by regulation to "take healthcare into your own hands" even if you want to.
It's a more-or-less intentional equivocation between different meanings of AI, as you note, machine learning vs generative AI. They want to point at the real but unsexy potential of ML for medical use in order to pump up the perceived value of LLMs. They want to imply to the general public and investors that LLMs are going to cure cancer.
Obviously still double check things, but it was moment of clarity I hadn't really had before this. Still needed the doctor and all the experience to diagnose and fix things, but relaying that info back to me is something doctors are only okay at. Try it out! take a summary sheet of a recent visit or incident and feed it in.
For instance, as a SWE, I get just a little help with boilerplate from the AI. I could usually have done it better, but sometimes the ask is both simple enough and boring enough that the code from the LLM actually produces something very close to what I would produce.
On the other side of the coin, a non-technical person using AI would be unable to properly understand and review the output.
Where it shines is on things that I am OK at. Like writing marketing copy. I can get by myself, but its slightly outside of my wheelhouse, but as long as I have a solid understanding of the product I can use AI to compliment my beginner/intermediate skills and produce something better than I would produce on my own.
A similar thing is writing tutorials. I write some code and documentation, but the tutorials are enough of a slog that I get distracted by my distaste for it. This is a good fit for AI.
I think this is where we will see AI help the most. Where someone's skillset includes the task at hand but at a secondary level where the user might doubt themselves or get distracted with the misery the task brings them.
If the proverbial marketer that you were referring to had some experience with coding, I dont see why they wouldnt be able to review the output and see any obvious flaws.
My whole point is that LLMs are of limited use when you are already an expert or when you know nothing about the subject. However, they really seem to help elevate beginner/intermediate level tangential skillsets.
Obviously everything is still evolving and your results may vary.
WT actual F? They invested so much into something what is not obvious brings value? Will there be consequences on them? Or they take the bonus and hide in New Zealand bunker?
It's big money betting on narratives from wanna-be-big money how AI is transformative for the future. Public takes all the risks with hardware and energy inflation or bailing out banks out of investments which require pruductivity growth from AI which we don't yet see in statistics.
We took the wrong turn somewhere. And responsible people don't seem to be capable or willing to change the course. Too much power in too few weak minds. Nothing good will come from this.
That’s courageous from a CEO of an US company, where the current government doesn’t see burning more oil as being bad for the planet, and is willing to punish everyone who thinks otherwise.
Cause they are able to search the web deeply, search for up to date info/research and synergize all that. You can have back and fourth for as long as you need.
The issue is that using LLMs properly requires a certain skill that more people lack.
And I don't mean I've just rejected the lowest offer to DIY shit to oblivion. I've accepted bids and been continuously disappointed with the results. To the point I no longer trust "experts" in these spaces because any "expertise" they bring is pretty shallow at best. The exception I'd point out is the auto-shop I prefer. They are busy enough that if I want to schedule something it's going to be 3-6 months out. As a result I've replaced my own suspension and replaced my alternator and starter myself while waiting for appointments with the rare actual experts in any domain. Actual expertise is rare and most folks don't know how to recognize it, especially outside of domains they are familiar with. Unfortunately it's way more profitable to fake expertise in various domains and collect payments and run than to actually stand behind your business and work. Thus the world we're forced to live in today.
This probably has nothing to do with gen AI (the kind of AI Nadela is speaking about).
Though it's a use case people like Satya will want to avoid for reasons.
LLMs will be used for aggressive, yet incredibly subtle manipulation, consensus building, and response tracking.
20-40% of social media is already bots, and in the future it is likely you will not be able to reply to anything anywhere without a bot either 1) responding, or 2) logging and sending your response to multiple parties instantly.
If the Stasi had LLMs the Berlin Wall would have never fallen
* Higher electricty bills.
* 5-6x cost of RAM, GPUs, and other computer components
* Data centers popping up in their backyards
* An internet inundated with slop
* Slop beginning to infiltrate the video game industry and other creative industries
* AI being used to justify gutting entry level jobs for a generation already screwed by larger, long horizon economic forces
* Grok enabling the creation of revenge porn and CSAM with seemingly no repercussions
* Massive IP theft on a scale previously unheard of
* Etc.
The pros of AI are:
* It can summarize text and transcribe audio decently well.
* It can make funny pictures of cats wearing top hats.
* ???
And no, I'm not saying the technology is bad. The business isn't going swimmingly, though.
There obviously are some compelling use cases for "AI", but it's certainly questionable if any of those are really making people's lives any better, especially if you take "AI" to mean LLMs and fake videos, not more bespoke uses like AlphaFold which is not only beneficial, but also not a resource hog.
There are plenty of uses for AI. Right now, the industry is heavily spending on training new models, improving performance of existing software and hardware, and trying to create niche products.
Power usage for inference will drop dramatically over the next decade, and more models are going to run on-device rather than in the cloud. AI is only going to become more ubiquitous, there's 0% chance it 'fails' and we return to 2020.
Only because companies have been cutting costs for decades here. This is not a good argument for AI.
> writing software
If you mean typing characters quickly, yes. Otherwise, there’s still a lot of employed devs, with many AI companies hiring.
> writing docs about software
The most useful docs are there because they contain info you cannot determine from the code. AI is not able to do this.
> computer graphics (animation, images)
If you are producing slop, yes.
> driving cars
True, but only because of its improved physical awareness. ie it’s a mechanical gain (better eyes, ears, etc) not an intellectual one (interpreting that information). Self driving cars aren’t LLMs and not really applicable here. Entirely different field.
> AI is only going to become more ubiquitous, there's 0% chance it 'fails' and we return to 2020
Absolutely true. But not for the reasons you think.
An AI might be better than an indian call center but I doubt that when the AI is made by indians anyway.
> writing software, writing docs about software
I have asked AI about exactly one topic and it lied about the API of a library making up the functions I was supposed to call.
> computer graphics (animation, images)
I have indeed seen many wonderful meme images come out of the generators but that was before they got lobotomized for producing that subject matter
[EDIT]
And the worst part is these are all just more "software as a service" designed to remove the possibility of using a tool without approval.
I think there are business reasons why they wouldn’t do that, and that makes me sad.
Every time it hallucinates visits to Starbucks.
I never go to Starbucks, it’s just a probable finding given the words in the question.
This should work. I want it to work. But until it can do this correctly all analysis capabilities should be suspect.
Even a year ago I had success with Claude giving it a photo of my credit card bill and asking it to give me repeating category subtotals, and it flawlessly OCR'd it and wrote a Python program to do as asked, giving me the output.
I'd imagine if you asked it to do a comparison to something else it'd also write code to do it, so get it right (and certainly would if you explicity asked).
When non techie friends/family bring up AI there are two major topics: 1) the amount of slop is off the charts and 2) said slop is getting harder to recognize which is scary. Sometimes they mention a bit of help in daily tasks at work, but nothing major.
They don't find AI useful, just a toy. Is their fault? Maybe.
Hi there, friends from another dimension! In my reality, there's a cold front coming from the north. Healthcare is expensive and politics are a mess. But AI? It hallucinates sometimes but it's so much better for searching, ad hoc consultation and as a code assistant than anything I've ever seen. It's not perfect, but it saved me SO much time I decided to pay for it. I'm a penny pincher, so I wouldn't be paying for it otherwise.
I think Satya is talking about cost/benefit. AI is incredibly useful but also incredibly expensive. I think we still need to find the right balance (perhaps slower model releases), but there's no way we'll put the genie back in the bottle.
I hope your AI gets better! Talk to you later!
I have access to all the popular AI tools from work for free, I use them for the same cases you mentioned like search, consultation, a better StackOverflow, and autocomplete. It’s definitely useful but I would describe that as incrementally useful, not revolutionary.
Satya is saying that AI needs to start doing more than vibe coding and autocomplete, there’s probably half a trillion invested into the technology worldwide now and it’s not enough for AI to be a good coding assistant. It needs to replace customer support, radiologists, and many other professions to justify the unprecedented level of investment its garnered.
AI is subsidized for the users
[0]: https://www.bloomberg.com/graphics/2025-ai-data-centers-elec...
So you can easily add 1-200 dollars to your bill for one day of higher usage.
https://www.myhorrynews.com/news/horry-electric-co-op-to-cha...
Copilot Notepad.
Copilot MS Paint.
Copilot Shoes.
Copilot Ice Cream.
Copilot Liposuction Surgery Machines.
LOL. "Looks like you're trying to tie those laces - would you like me to order you velcro?"
Might not work well with those relying on structured finance to break computer component markets, but does well for the majority that are buying their equipment with regular cash.
With all this useless slop, he’s literally arguing against his own point.
No brainwr.
And yet studies show the opposite [0].
[0] https://www.media.mit.edu/publications/your-brain-on-chatgpt...
His bottom line depends on this bet that everyone is going to depend on AI and pay Microslop rent to use it.
https://scholar.google.ca/scholar?q=cognitive+effects+of+ai+...
I’ve been predicting for a while: free or cheap AI will enshittify and become an addictive ad medium with nerfed capabilities. If you want actually good AI you will have to pay for it, either a much heftier fee or buying or renting compute to run your own. In other words you’ll be paying what it actually costs, so this is really just the disappearance of the bubble subsidy.
So this is something that factors in hugely in planning permission. What do we get back for it is a question asked a lot. And datacenters are notoriously bad at providing jobs, during construction yes but in the run phase it's mainly low-value remote hands and security stuff.
https://youtu.be/M0S3a32RzEo?t=278
But they aren't stupid. You sound like a tech bro.
That's the problem.