[HN Gopher] Microsoft is plotting a future without OpenAI
___________________________________________________________________
Microsoft is plotting a future without OpenAI
Author : doublebind
Score : 238 points
Date : 2025-03-07 18:44 UTC (4 hours ago)
(HTM) web link (techstartups.com)
(TXT) w3m dump (techstartups.com)
| doublebind wrote:
| Original story: Microsoft's AI Guru Wants Independence From
| OpenAI. That's Easier Said Than Done,
| https://www.theinformation.com/articles/microsofts-ai-guru-w...
| mirekrusin wrote:
| I don't get this "easier said than done" part.
|
| There are really not that many things in this world you can
| swap as easily as models.
|
| Api surface is stable and minimal, even at the scale that
| microsoft is serving swapping is trivial compared to other
| things they're doing daily.
|
| There is enough of open research results to boost their phi or
| whatever model and be done with this toxic to humanity, closed,
| for profit company.
| jsemrau wrote:
| For cloud providers it makes sense to be model agnostic.
|
| While we still live in a datacenter driven world, models will
| become more efficient and move down the value chain to consumer
| devices.
|
| For Enterprise, these companies will need to regulate model risk
| and having models fine-tuned on proprietary data at scale will be
| an important competitive differentiator.
| aresant wrote:
| Thematically investing billions into startup AI frontier models
| makes sense if you believe in first-to-AGI likely worth a
| trillion dollars +
|
| Investing in second/third place likely valuable at similar scales
| too
|
| But outside of that MSFTs move indicates that frontier models
| most valuable current use case - enterprise-level API users - are
| likely to be significantly commoditized
|
| And likely majority of proceeds will be captured by (a) those
| with integrated product distribution - MSFT in this case and (b)
| data center partners for inference and query support
| j45 wrote:
| First to AGI for the big companies? Or for the masses?
|
| Computationally, some might have access to it earlier before
| it's scalable.
| Retric wrote:
| Profit from say 3 years of enterprise AGI exclusivity is
| unlikely to be worth the investment.
|
| It's moats that capture most value not short term profits.
| alabastervlog wrote:
| At this point, I don't see much reason to believe the "AGI is
| imminent and these things are potentially dangerous!" line at
| all. It looks like it was just Altman doing his thing where he
| makes shit up to hype whatever he's selling. Worked great, too.
| "Oooh, it's _so_ dangerous, we're so concerned about safety!
| Also, you better buy our stuff."
| torginus wrote:
| but all those ominous lowercase tweets
| only-one1701 wrote:
| What even is AGI? Like, what does it look like? Genuine
| question.
| taneq wrote:
| It's whatever computers can't do.
| lwansbrough wrote:
| An AI agent with superhuman coherence that can run
| indefinitely without oversight.
| only-one1701 wrote:
| People sincerely think we're < 5 years away from this?
| jimbokun wrote:
| Is there some fundamental constraint keeping it from
| happening? What cognitive capability do humans have that
| machines won't be able to replicate in that time frame?
|
| Each remaining barrier has been steadily falling.
| bigstrat2003 wrote:
| We don't even have AI which can do useful things yet. The
| LLMs these companies make are fun toys, but not useful
| tools (yes, I know that hype-prone people are using them
| as such regardless). It beggars belief that we will go
| from "it's a fun toy but can't do real work" to "this can
| do things without even needing human supervision" without
| a major leap in capabilities.
| taco_emoji wrote:
| What barriers have fallen? Computers still can't even
| drive cars
| bobsmooth wrote:
| Even with cutting edge technology the number of
| transistors on a chip is nowhere close to the number of
| neurons in the brain.
| saint_yossarian wrote:
| Creativity, tastes, desires?
|
| All the LLM tech so far still requires a human to
| actually prompt them.
| bashfulpup wrote:
| Continual Learning, it's a barrier that's been there from
| the very start and we've never had a solution to it.
|
| There are no solutions even at the small scale. We
| fundamentally don't understand what it is or how to do
| it.
|
| If you could solve it perfectly on Mnist just scale and
| then we get AGI.
| Spooky23 wrote:
| People on HN in 2015 were saying that by now car
| ownership would be dying and we'd be renting out our self
| driving cars as we sat at work and did fuck all. Ben
| Thompson had podcasts glazing Uber for 3 hours a month.
|
| The hype cycle for tech people is like a light bulb for a
| moth. We're attracted to potential, which is both our
| superpower and kryptonite.
| valiant55 wrote:
| Obviously the other responder is being a little tongue-in-
| cheek but AGI to me would be virtually indistinguishable from
| a human in both ability to learn, grow and adapt to new
| information.
| Enginerrrd wrote:
| Honestly it doesn't even need to learn and grow much if at
| all if its able to properly reason about the world and its
| context and deal with the inexhaustible supply of
| imperfections and detail with reality.
| bashfulpup wrote:
| That implies learning. Solve continual learning and you
| have agi.
|
| Wouldn't it amaze you if you learned 10 years ago that we
| would have AI that could do math and code better than 99%
| of all humans. And at the same time they could barely
| order you a hotdog on doordash.
|
| Fundamental ability is lacking. AGI is just as likely to
| be solved by Openai as it is by a college student with a
| laptop. Could be 1yr or 50yrs we cannot predict when.
| Enginerrrd wrote:
| Strictly speaking I'm not sure if it does require
| learning if information representing the updated context
| is presented. Though it depends what you define as
| learning. ("You have tried this twice, and it's not
| working.") is often enough to get even current LLM's to
| try something else.
|
| That said, your second paragraph is one of the best and
| most succinct ways of pointing out why current LLM's
| aren't yet close to AGI if though they sometimes feel
| like it's got the right idea.
| samtp wrote:
| Would it also get brainrot from consuming too much social
| media & made up stories? Because I imagine it's reasoning
| would have to be significantly better than the average
| human to avoid this.
| ge96 wrote:
| Arnold, a killing machine that decides to become a handy man
|
| Zima blue was good too
| zombiwoof wrote:
| I'm here to fix the cable
|
| Logjammin AI
| c0redump wrote:
| A machine that has a subjective consciousness, experiences
| qualia, etc.
|
| See Thomas Nagels classic piece for more elaboration
|
| https://www.sas.upenn.edu/~cavitch/pdf-library/Nagel_Bat.pdf
| myhf wrote:
| The official definition of AGI is a system that can generate
| at least $100 billion in profits. For comparison, this would
| be like if perceptrons in 1968 could generate $10 billion in
| profits, or if LISP machines in 1986 could generate $35
| billion in profits, or if expert systems in 1995 could
| generate $50 billion in profits.
| mirekrusin wrote:
| Apparently according to ClosedAI it's when you charge for API
| key the same as salary for employee.
| coffeefirst wrote:
| It's the messiah, but for billionaires who hate having to pay
| people to do stuff.
| lm28469 wrote:
| Short term betting on AGI from current LLMs is like if you
| betted on V10 F1s two weeks after we invented the wheel
| oezi wrote:
| Not the worst bet to invest in Daimler when they came up with
| the car. Might not get you to F1, but certainly a good bet
| they might.
| laluser wrote:
| I think they both want a future without each other. OpenAI will
| eventually want to vertically integrate up towards applications
| (Microsoft's space) and Microsoft wants to do the opposite in
| order to have more control over what is prioritized, control
| costs, etc.
| Spooky23 wrote:
| I think OpenAI is toxic. Weird corporate governmance shadiness.
| The Elon drama, valuations based on claims that seem like the
| AI version of the Uber for X hype of a decade ago (but
| exponentially crazier). The list goes on.
|
| Microsoft is the IBM of this century. They are conservative,
| and I think they're holding back -- their copilot for
| government launch was delayed months for lack of GPUs. They
| have the money to make that problem go away.
| skinnymuch wrote:
| IBM of this century in a good way?
| optimalsolver wrote:
| IBM of the early 1940s.
| Spooky23 wrote:
| In this context, it's not good or bad, it just is.
| bredren wrote:
| Despite the actual performance and product implementation, this
| suggests to me Apple's approach was more strategic.
|
| That is, integrating use of their own model, amplifying
| capability via OpenAI queries.
|
| Again, this is not to drum up the actual quality of the product
| releases so far--they haven't been good--but the foundation of
| "we'll try to rely on our own models when we can" was the right
| place to start from.
| strangescript wrote:
| I think they have realized that even if OpenAI is first, it won't
| last long so really its just compute at scale, which is something
| they already do themselves.
| echelon wrote:
| There is no moat in models (OpenAI).
|
| There is a moat in infra (hyperscalers, Azure, CoreWeave).
|
| There is a moat in compute platform (Nvidia, Cuda).
|
| Maybe there's a moat with good execution and product, but it
| isn't showing yet. We haven't seen real break out successes. (I
| don't think you can call ChatGPT a product. It has zero
| switching cost.)
| drumhead wrote:
| Is anyone other than Nvdia making money from this particular
| gold rush?
| xnx wrote:
| Data center construction and power companies.
| scarface_74 wrote:
| Consulting companies
| barumrho wrote:
| Given xAI built its 100k gpu datacenter in a very short time,
| is the infra really a moat?
| freedomben wrote:
| I'd say it is because the $ it takes to build out even a
| small gpu data center is still way, way more than most
| small cos can do. It's not an impenetrable moat, but it is
| pretty insulating against startups. Still have a threat
| from big tech, though I think that will always be true for
| almost everything
| eagerpace wrote:
| I don't think the hardware is that easy to source just yet.
| Musk pulled some strings and redirected existing inventory
| and orders from his other companies, namely Tesla, to
| accelerate delivery.
| PKop wrote:
| xAI does not have infra to sell the service and
| integrations of it to enterprises and such. It's an open
| question if "models" alone and simple consumer products
| that use them are profitable. So, probably hyperscale cloud
| platform infra is a moat yes. Microsoft has Semantic
| Kernel, Microsoft.Extensions.AI, various RAG and search
| services, and an entire ecosystem and platform around using
| LLM's to build with that xAI does not have. Just having a
| chat app as interface to one's model is part of the
| discussion here about models as commodities. xAI does have
| X/Twitter data which is a constantly updating source of
| information so in that aspect they themselves do have
| something unique.
| YetAnotherNick wrote:
| What moat does Nvidia have. AMD could have ROCm perfected if
| they really want to. Also most of pytorch, specially those
| relevant to transformers runs perfectly on Apple Silicon and
| TPUs and probably other hardware as well.
|
| If anyone has moat related to Gen AI, I would say it is the
| data(Google, Meta).
| klelatti wrote:
| > AMD could have ROCm perfected if they really want to.
|
| It's not an act of will or CEO dictat. It's about hiring
| and incentivising the right people, putting the right
| structures in place etc all in the face of competing
| demands.
|
| Nvidia have a huge head start and by the time AMD have
| 'caught up' Nvidia with it's greater resources will have
| moved further ahead.
| YetAnotherNick wrote:
| If head start is a moat, why wouldn't you count OpenAI's
| headstart as moat?
| echelon wrote:
| Anyone can make an LLM. There are hundreds of choices in
| the market today. Many of them are even open source.
|
| OpenAI brings absolutely nothing unique to the table.
| klelatti wrote:
| Because we already see firms competing effectively with
| OpenAI.
|
| There is as yet no indication that AMD can match Nvidia's
| execution for the very good reason that doing so is
| extremely difficult. The head start is just the icing on
| the cake.
| PKop wrote:
| Not all industries or product segments are equal is the
| obvious answer. The point here whether one agrees or not
| is models are easier to catch up to than GPUs
| kittikitti wrote:
| Surprising how Sam Altman's firing as CEO of OpenAI and moving to
| Microsoft wasn't mentioned in this article.
| electriclove wrote:
| Do you have a source?
| selimthegrim wrote:
| They mean the past events.
| DeathArrow wrote:
| It's only logical. OpenAI it's too expensive for what it
| produces. Deep Seek is on par with ChatGPT and the cost was
| lower. Claude development costs less, too.
| knowitnone wrote:
| Good. I'm plotting a future without Microsoft
| meepmeepinator wrote:
| Microsoft's shift away from OpenAI reminds me of Google's early
| AI struggles. Back in 2016, Google relied heavily on Nvidia GPUs
| for training models but saw the long-term cost risk. So, they
| built TPUs--custom AI chips--to take control of their
| infrastructure. Now, Microsoft is doing the same: developing in-
| house AI models (Phi-4) and custom silicon (Maia) to reduce
| reliance on OpenAI and Nvidia. But history shows that model
| independence is harder than it looks. Microsoft's models are
| promising, but GPT-4 still outperforms them in general tasks.
| Meanwhile, integrating multiple models (OpenAI, Meta, Anthropic)
| into 365 Copilot is a major engineering challenge--consistency
| and latency issues are inevitable. If they pull it off, they'll
| transform Azure into an AI-agnostic powerhouse. If not, they risk
| fragmentation and higher costs. Either way, this move signals the
| next phase of AI competition: infrastructure control.
| mattlondon wrote:
| Why not just use GCP? It is already model agnostic
| https://console.cloud.google.com/vertex-ai/model-garden
|
| There is even deepseek on there.
| agentultra wrote:
| I had skimmed the headline and thought, "Microsoft is plotting a
| future without AI," and was hopeful.
|
| Then I read the article.
|
| Plotting for a future without Microsoft.
| mirekrusin wrote:
| First quarter summary of this year is "AI is plotting future
| without OpenAI or Microsoft".
| CodeCompost wrote:
| Just partner with Deepseek
| Frederation wrote:
| Why.
| keernan wrote:
| From the article:
|
| Suleyman's team has also been testing alternatives from
| companies like xAI, DeepSeek, and Meta
| rdtsc wrote:
| They probably saw the latest models like gpt 4.5 not being as
| revolutionary as expected and deepseek and others catching up.
| thewebguyd wrote:
| I think Microsoft isn't buying the AGI hype from OpenAI, and
| wants to move to be more model agnostic, and instead do what
| Microsoft (thinks) it does best, and that's tooling, and
| enterprise products.
|
| MS wants to push Copilot, and will be better off not being tied
| to OpenAI but having Copilot be model agnostic, like GH Copilot
| can use other models already. They are going to try and
| position Azure as "the" place to run your own models, etc.
| rdtsc wrote:
| > instead do what Microsoft (thinks) it does best, and that's
| tooling, and enterprise products.
|
| Definitely, but I think it's because they saw OpenAI's moat
| get narrower and shallower, so to speak. As the article
| mentions it's still looking like a longer timeline [quote]
| "but Microsoft still holds exclusive rights to OpenAI's
| models for its own products until 2030. That's a long
| timeline to unravel."
| only-one1701 wrote:
| Maybe I'm just cynical, but I wonder how much of this initiative
| and energy is driven by people at Microsoft who want their own
| star to rise higher than it can when it's bound by a third-party
| technology.
|
| I feel like this is something I've seen a fair amount in my
| career. About seven years ago, when Google was theoretically
| making a big push to stage Angular on par with React, I remember
| complaining that the documentation for the current major version
| of Angular wasn't nearly good enough to meet this stated goal. My
| TL at the time laughed and said the person who spearheaded that
| initiative was already living large in their mansion on the hill
| and didn't give a flying f about the fate of Angular now.
| skepticATX wrote:
| Listening to Satya in recent interviews I think makes it clear
| that he doesn't really buy into OpenAI's religious-like view of
| AGI. I think the divorce makes a lot of sense in light of this.
| keeganpoppen wrote:
| oh it is absolutely about that
| bsimpson wrote:
| There is a prominent subset of the tech crowd who are ladder
| climbers - ruthlessly pursuing what is rewarded with
| pay/title/prestige without regard to actually making good
| stuff.
|
| There are countless kidding-on-the-square jokes about projects
| where the innovators left at launch and passed it off to the
| maintenance team, or where a rebrand was in pursuit of
| someone's promo project. See also, killedbygoogle.com.
| teaearlgraycold wrote:
| These people should be fired. I want a tech company where
| people are there to make good products first and get paid
| second. And the pay should be good. The lifestyle
| comfortable. No grindset bullshit. But I am confident that if
| you only employ passionate people working their dream jobs
| you will excel.
| escapecharacter wrote:
| Unfortunately whether someone is checked out is a laggy
| measure.
|
| Even good honest motivated people can become checked out
| without even being aware of it.
|
| The alternative is to lay off people as soon as they hit
| 1.0 (with a severance bonus on the scale of an
| acquisition). This would obviously be worse, as you can't
| take advantage of their institutional knowledge.
| saturn8601 wrote:
| This motivated part of Musk's moves at Twitter(and now
| DOGE). You can't reliably evaluate which people are
| checked out and when you are against the clock, you have
| to take a hatchet and accept that you will break things
| that are in motion.
| scarface_74 wrote:
| Why would those people be "fired" when the entire promotion
| process and promo docs emphasize "scope" and "impact"?
|
| No one works for any BigTech company because they think
| they are making the world a better place. They do it
| because a shit ton of money appears in their bank account
| every pay period and stock appears in their brokerage
| account every vesting period.
|
| I personally don't have the shit tolerance to work in
| BigTech (again) at 50. But I suggest to all of my younger
| relatives who graduate in CS to "grind leetCode and work
| for a FAANG" and tell them how to play the politics to get
| ahead.
|
| As the Dilbert author said, "Passion is Bullshit". I have
| never been able to trade passion for goods and services.
| bsimpson wrote:
| > No one works for any BigTech company because they think
| they are making the world a better place.
|
| I'm sure there are plenty of people who work at big
| companies for precisely this reason (or at least, with
| that as _a_ reason among many).
|
| Yes, much of the prestige has worn off as the old guard
| retired and current leadership emphasizes chasing AI
| buzzwords and cutting costs. But still, big companies are
| one of the few places where an individual really can
| point out something they worked on in day-to-day life.
| (Pull out any Android phone and I can show you the parts
| that my work touched.)
| Severian wrote:
| Funny what his passions turned into, so yeah, ironically
| agree.
| whstl wrote:
| Yep. I've seen more people fired for being passionate
| about their craft and their jobs than people getting
| raises for the same reason.
|
| It's always the same. People trying to make things better
| for the next developer, people prioritizing delivers
| instead of ego-projects or ego-features by someone
| playing politics, developers wanting a seat at the table
| with (dysfunctional) Product teams, people actual good
| intentions trying to "change the world" (not counting the
| misguided attempts here).
|
| You are 100% correct, you gotta play the politics,
| period.
| JumpCrisscross wrote:
| > _want a tech company where people are there to make good
| products first and get paid second. And the pay should be
| good. The lifestyle comfortable. No grindset bullshit_
|
| Congratulations, you've invented the HR department in
| corporate America.
| saturn8601 wrote:
| You are trying to combine two repelling magnets together.
|
| Case in point: Tesla/SpaceX meets your first criteria: "I
| want a tech company where people are there to make good
| products first and get paid second."
|
| Google meets your second criteria: "And the pay should be
| good. The lifestyle comfortable. No grindset bullshit."
|
| Other than small time boutique software firms like Fog
| Creek Software or Panic Inc(and thats a BIG maybe) you are
| not going to get this part of your message: "But I am
| confident that if you only employ passionate people working
| their dream jobs you will excel."
|
| There are tradeoffs in life and each employee has to choose
| what is important to them(and each company CEO has to set
| standards on what is truly valued at the company).
| supriyo-biswas wrote:
| At my former employer, there was a team who were very much
| into resume-driven development and wrote projects in Go even
| when Java would have been the better alternative considering
| the overall department and maintenance team expertise, all
| the while they were informally grumbling about how Go doesn't
| have the features they need...
| darkhorse222 wrote:
| I see that a lot from the Go crowd. That's why I consider
| any strong opinions on languages to be a poor indicator for
| ability. Sure there's differences, but a language does not
| make the engineer. Someone who is attracted to flashy stuff
| makes for an indulgent planner.
| scubbo wrote:
| > That's why I consider any strong opinions on languages
| to be a poor indicator for ability.
|
| Hmm. Can't say I agree here - at least not with the
| literal text of what you've written (although maybe we
| agree in spirit). I agree that _simplistic_ strong
| opinions about languages are a sign of poor
| thoughtfulness ("<thing> is good and <other thing> is
| bad") - but I'd very much expect a Staff+ engineer to
| have enough experience to have strong opinions about the
| _relative_ strengths of various languages, where they're
| appropriate to use and where a different language would
| be better. Bonus points if they can tell me the worst
| aspects about their favourite one.
|
| Maybe we're using "opinion" differently, and you'd call
| what I described there "facts" rather than opinions. In
| which case - yeah, fair!
| mikepurvis wrote:
| Absolutely. Anyone senior should be able to fairly
| quickly get a handle on the requirements for a particular
| project and put forward a well-reasoned opinion on an
| appropriate tech stack for it. There might be some blank
| space in there for "I've heard of X and Y that actually
| might fit this use case slightly better, so it's probably
| worth a brief investigation of those options, but I've
| used Z before so I know about the corner cases we may run
| into, and that has value too."
| pdimitar wrote:
| And I see people who assume choosing a language was done
| for "flashy stuff" the less capable.
|
| See, we can all generalize. Not productive.
|
| Only thing I ever saw from Golang devs was pragmatism. I
| myself go either for Elixir or Rust and to me Golang sits
| in a weird middle but I've also written 20+ small tools
| for myself in Golang and have seen how much quicker and
| more productive I was when I was not obsessed with
| complete correctness (throwaway script-like programs,
| small-to-mid[ish]-sized projects, internal tools etc.)
|
| You would do well to stop stereotyping people based on
| their choice of language.
| zozbot234 wrote:
| > how much quicker and more productive I was when I was
| not obsessed with complete correctness
|
| That's pretty much another way of saying that stuff
| becomes a whole lot quicker and easier when you end up
| getting things wrong. Which may even be true, as far as
| it goes. It's just not very helpful.
| pdimitar wrote:
| Obviously. But I did qualify my statement. There are
| projects where you're OK with not getting everything
| right from the get go.
|
| FWIW I very much share your exact thoughts on Rust
| skewing metrics because it makes things too easy and
| because stuff almost immediately moves to maintenance
| mode. But that being said, we still have some tasks where
| we need something yesterday and we can't argue with the
| shot-callers about it. (And again, some personal projects
| where the value is low and you derive more of it if you
| try quickly.)
| BobbyJo wrote:
| Language matters quite a bit when deciding how to build
| an application though. I see having no strong opinions on
| language to be a sign the person hasn't developed a wide
| enough variety of projects to get a feel for their
| strengths and weaknesses.
| ohgr wrote:
| Yeah. I have a list of things I won't work with. That's
| what experience looks like.
|
| (Mostly .Net, PHP and Ruby)
| synergy20 wrote:
| golang is decent and is the only new lang climbed up to 7th
| in popularity, it does shine at what it's good at
| mvdtnz wrote:
| You're missing the point completely.
| pclmulqdq wrote:
| The go and rust crowds both love writing things in their
| own language for its own sake. Not because it's a good
| choice. For a large web backend, go is great. For many
| other things it's terrible.
| pdimitar wrote:
| > _The go and rust crowds both love writing things in
| their own language for its own sake_
|
| Hard to take you seriously when you do such weird
| generalized takes.
|
| While it's a sad fact that fanboys and zealots absolutely
| do exist, most devs can't afford to be such and have to
| be pragmatic. They pick languages based on merit and
| analysis.
| pclmulqdq wrote:
| Most of the people who use Go and Rust do it for
| pragmatic reasons. That doesn't influence the culture of
| the zealots in each community.
|
| You should search for headlines on HN that say "written
| in Go" or "written in Rust" and then compare that to the
| number of headlines that say "written in JavaScript" or
| "written in Kotlin."
| zozbot234 wrote:
| Rust is pretty antithetical to resume-driven development
| because a lot of the stuff that's written in Rust is
| _too_ well-written and free of the usual kinds of
| software defects. It immediately becomes "done" and
| enters low-effort maintenance mode, there's just very
| little reason to get involved with it anymore since "it
| just works". Believe it or not, this whole dynamic is
| behind a lot of the complaints about Rust in the
| workplace. It's literally making things _too_ easy.
| ohgr wrote:
| Having watched two entirely fucked up Rust projects get
| written off I think you need to get out more.
| LPisGood wrote:
| It's not that I don't believe you, hut that I'm having
| trouble seeing how what you say could be true.
|
| Rust projects immediately become "done"??? They don't
| also having changing requirements and dependencies? Why
| aren't everyone at the best shops using it for everything
| if it massively eliminates work load?
| pclmulqdq wrote:
| I have to say that the median crate I interact with has
| the following readme:
|
| " Version 0.2 - Unstable/buggy/slow unless you use
| exactly like the example - not going to get updated
| because I moved on to something else"
|
| Rust is another programming language. It's easier to
| write code without a certain class of bugs, but that
| doesn't mean version 0.2 of a casual project is going to
| be bug-free.
| conjectures wrote:
| Not from what I've seen. The compiler is slow af which
| plays badly with how fussy the thing is.
|
| It's easy to have no defects in functionality you never
| got around to writing because you ran out of time.
| whstl wrote:
| Life is too short to program on languages one doesn't
| love.
|
| Those people, if they really exist, are right.
| rurp wrote:
| I've seen the exact same pattern play out with different
| tools. The team used a shiny new immature platform for nice
| sounding reasons and then spent 80% of their time
| reinventing wheels that have already been solved in any
| number of places.
| whstl wrote:
| I have lot of sympathy for resume-driven developers.
| They're just answering to the labor market. More power to
| them.
|
| When companies do what the market expect we praise them.
| When it's workers, we scorn them. This attitude is
| seriously fucked up.
|
| When companies start hiring based on experience,
| adaptability, curiosity, potential and curiosity then you
| get to complain. Until that, anyone doing it should be
| considered a fucking genius.
| usefulcat wrote:
| Pretty sure most of the resentment comes from working
| with such people. Which I think is understandable.
| whstl wrote:
| Understandable, but still wrongfully blaming a player
| rather than the game itself.
| fallingknife wrote:
| What good does that do on a resume? I thought learning a
| new language on the job was pretty standard.
| ohgr wrote:
| We have those! Turn up, make some micro-services or AWS
| crap pile we don't need to solve a simple problem, then
| fuck off somewhere else and leave everyone else to clean it
| up.
|
| Worst one is the data pipeline we have. It's some AWS
| lambda mess which uses curl to download a file from
| somewhere and put it into S3. Then another lambda turns up
| at some point and parses that out and pokes it into
| DynamoDB. This fucks up at least once a month because the
| guy who wrote the parser uses 80s BASIC style string
| manipulation and luck. Then another thing reads that out of
| DynamoDB and makes a CSV (sometimes escaped improperly) and
| puts that into another bucket.
|
| I of course entirely ignore this and use one entire line of
| R to do the same job
|
| Along comes a senior spider and says "maybe we can fix all
| these problems with AI". No you can stop hiring acronym
| collectors.
| conjectures wrote:
| Ah, the good ole Rube Goldberg machine.
| Lerc wrote:
| I had not encountered the phrase kidding-on-the-square
| before. Searching seems to reveal a spectrum of opinions as
| to what it means. It seems distinct to the 'It's funny
| because it's true' of darker humour.
|
| It seems to be more on a spectrum of 'Haha, only joking'
| where the joke teller makes a statement that is ambiguously
| humorous to measure the values of the recipients, or if they
| are not sure of the values of the recipients.
|
| I think the distinction might be on whether the joke teller
| is revealing (perhaps unintentionally) a personal opinion or
| whether they are making an observation on the world in
| general, which might even imply that they hold a counter-
| opinion.
|
| Where do you see 'kidding on the square' falling?
|
| (apologies for thread derailment)
| MWil wrote:
| good god, lemon
| bsimpson wrote:
| It's a phrase I learned from my mom/grandpa growing up. "On
| the square" means "but I also mean it."
| gsf_emergency_2 wrote:
| https://hn.algolia.com/?dateRange=pastWeek&page=0&prefix=
| tru...
| devsda wrote:
| > There is a prominent subset of the tech crowd who are
| ladder climbers - ruthlessly pursuing what is rewarded with
| pay/title/prestige without regard to actually making good
| stuff.
|
| I think the hiring and reward practices of the organizations
| & the industry as a whole also encourages this sort of
| behavior.
|
| When you reward people who are switching too often or only
| when moving internally/externally, switching becomes the
| primary goal and not the product. If you know beforehand that
| you are not going to stay long to see it through, you tend to
| take more shortcuts and risks that becomes the responsibility
| of maintainers later.
|
| We have a couple of job hoppers in our org where the number
| of jobs they held is almost equal to their years of
| experience and their role is similar to those with twice the
| experience! One can easily guess what their best skill is.
| deadbabe wrote:
| This is wrong.
|
| Google kills off projects because the legal liability and
| security risks of those projects becomes too large to justify
| for something that has niche uses or gives them no revenue.
| User data is practically toxic waste.
| grepLeigh wrote:
| As an outsider looking at Microsoft, I've always been
| impressed by the attention to maintaining legacy APIs and
| backward compatibility in the Windows ecosystem. In my mind,
| Microsoft is at the opposite end of the killedbygoogle.com
| spectrum. However, none of this is grounded in real evidence
| (just perception). Red Hat is another company I'd put forth
| as an example of a long-term support culture, although I
| don't know if that's still true under IBM.
|
| I'd love to know if my superficial impression of Microsoft's
| culture is wrong. I'm sure there's wild variance between
| organizational units, of course. I'm excluding the Xbox/games
| orgs from my mental picture.
| mlazos wrote:
| One of my friends stated this phenomenon very well "it's a
| lever they can pull so they do it". Once you've tied your
| career to a specific technology internally, there's really only
| one option: keep pushing it regardless of any alternatives
| because your career depends on it. So that's what they do.
| roland35 wrote:
| Unfortunately I don't think there is any real metric-based way
| to prevent this type of behavior, it just has to be old
| fashioned encouraged from the top. At a certain size it seems
| like this stops scaling though
| ambicapter wrote:
| Does it not make sense to not tie your future to a third-party
| (aka build your business on someone else's platform)? Seems
| like basic strategy to me if that's the case.
| pphysch wrote:
| It's a good strategy. It should be obvious to anyone paying
| attention that OpenAI doesn't have AGI secret sauce.
|
| LLMs are a commodity and it's the platform integration that
| matters. This is the strategy that Google, Apple embraced and
| now Microsoft is wisely pivoting to the same.
|
| If OpenAI cares about the long-term welfare of its employees,
| they would beg Microsoft to acquire them outright, before the
| markets fully realize what OpenAI is not.
| Izikiel43 wrote:
| > now Microsoft is wisely pivoting to the same.
|
| I mean, they have been doing platform integration for a
| while now, with all the copilot flavors and teams
| integrations, etc. This would change the backend model to
| something inhouse.
| pradn wrote:
| It's the responsibility of leadership to set the correct goals
| and metrics. If leadership doesn't value maintenance, those
| they lead won't either. You can't blame people for playing to
| the tune of those above them.
| ewhanley wrote:
| This is exactly right. If resume driven development results
| in more money, people are (rightly) going to do it. The
| incentive structure isn't set by the ICs.
| m463 wrote:
| I wonder if incentives for most companies favor doing things
| in-house?
| esafak wrote:
| Yes, you can say you built it from scratch, showing
| leadership and impact, which is what big tech promotions are
| gauged by.
| HarHarVeryFunny wrote:
| OpenAI already started divorce proceedings with their
| datacenter partnership with Softbank/etc, and it'd hardly be
| prudent for the world's largest software company NOT to have
| it's own SOTA AI models.
|
| Nadella might have initially been caught a bit flat footed with
| the rapid rise of AI, but seems to be managing the situation
| masterfully.
| wkat4242 wrote:
| In what world is what they are doing masterful? Their product
| marketing is a huge mess, they keep changing the names of
| everything every few months. Nobody knows which Copilot does
| what anymore. It really feels like they're scrambling to be
| first to market. It all feels so incredibly rushed.
|
| Whatever is there doesn't work half the time. They're hugely
| dependent on one partner that could jump ship at any moment
| (granted they are now working to get away from that).
|
| We use Copilot at work but I find it very lukewarm. If we
| weren't a "Microsoft shop" I don't think would have chosen
| it.
| aaronblohowiak wrote:
| > scrambling to be first
|
| Third?
| trentnix wrote:
| _> Their product marketing is a huge mess, they keep
| changing the names of everything every few months. Nobody
| knows which Copilot does what anymore. It really feels like
| they 're scrambling to be first to market. It all feels so
| incredibly rushed._
|
| Product confusion, inconsistent marketing, unnecessary
| product renames, and rushing half-baked solutions has been
| the Microsoft way for dozens of products across multiple
| divisions for years.
| eitally wrote:
| Rule #1 for Microsoft product strategy: if you can't
| yourselves figure out the SKUs and how they bundle
| together, the odds are good that your customers will
| overpay. It's worked for almost 50 years and there's no
| evidence that it will stop working. Azure is killing it
| and will continue to eat the enterprise even as AWS
| starts/continues to struggle.
| HarHarVeryFunny wrote:
| > In what world is what they are doing masterful?
|
| They got access to the best AI to offer to their customers
| on what seems to be very favorable terms, and bought
| themselves time to catch up as it now seems they have.
|
| GitHub Copilot is a success even if Microsoft/Windows
| Copilot isn't, but more to the point Microsoft are able to
| offer SOTA AI, productized as they see fit (not every
| product is going to be a winner) rather than having been
| left behind, and corporate customers are using AI via Azure
| APIs.
| nyarlathotep_ wrote:
| > In what world is what they are doing masterful?
|
| Does *anyone* want "Copilot integration" in random MS
| products?
| saturn8601 wrote:
| Ah man I don't want to hear things like that. I work in an
| Angular project and it is the most pleasant thing I have worked
| with (and i've been using it as my primary platform for almost
| a decade now). If I could, i'd happily keep using this
| framework for the rest of my career(27 years to go till
| retirement).
| hintymad wrote:
| > but I wonder how much of this initiative and energy is driven
| by people at Microsoft who want their own star to rise higher
| than it can when it's bound by a third-party technology.
|
| I guess it's human nature for a person or an org to own their
| own destiny. That said, the driving force is not personal
| ambition in this case though. The driving force behind this is
| that people realized that OAI does not have a moat as LLMs are
| quickly turning into commodities, if haven't yet. It does not
| make sense to pay a premium to OAI any more, let alone at the
| cost of not having the flexibility to customize models.
|
| Personally, I think Altman did a de-service to OAI by
| constantly boasting AGI and seeking regulatory capture, when he
| perfectly knew the limitation of the current LLMs.
| RobertDeNiro wrote:
| xAI could do it, deepseek could do it . Microsoft can as well.
| It's not hard to see
| rafaelmn wrote:
| I'd be willing to bet that the largest use of LLMs they have is
| GitHub copilot and Claude should be the default there.
|
| OpenAI has not been interesting to me for a long time, every time
| I try it I get the same feeling.
|
| Some of the 4.5 posts have been surprisingly good, I really like
| the tone. Hoping they can distill that into their future models.
| DebtDeflation wrote:
| A couple of days ago it leaked that OpenAI was planning on
| launching new pricing for their AI Agents. $20K/mo for their PhD
| Level Agent, $10K/mo for their Software Developer Agent, and
| $2K/mo for their Knowledge Worker Agent. I found it very telling.
| Not because I think anyone is going to pay this, but rather
| because this is the type of pricing they need to actually make
| money. At $20 or even $200 per month, they'll never even come
| close to breaking even.
| paxys wrote:
| It's pretty funny that OpenAI wants to sell access to a "PhD
| level" model at a price with which you can hire like 3-5 real
| human PhDs full-time.
| laughingcurve wrote:
| That is just not correct. As someone who has done the budgets
| for PhD hiring and funding, you are just wildly
| underestimating the overhead costs, benefits, cost of raising
| money, etc.
| zombiwoof wrote:
| Respectfully disagree. I had two pHD on a project and spent
| a total of 120k a year on them.
| 0_____0 wrote:
| What region and what field?
| yifanl wrote:
| Right, which is substantially less than the stated
| $20k/month.
|
| edit: I see we're actually in agreement, sorry, I read
| the indentation level wrong.
| robertlagrant wrote:
| > Respectfully disagree. I had two pHD on a project and
| spent a total of 120k a year on them.
|
| Does that include all overheads such as HR, payroll, etc?
| eszed wrote:
| How many PhDs can you afford for $20k a month in your
| field?
| DebtDeflation wrote:
| The "3-5" is certainly overstated, but you definitely can
| hire ONE PhD for that price, just as you can hire a SWE for
| $120K or a knowledge worker for $24K. The point is that
| from a CEO's perspective "replacing all the humans with AI"
| looks a lot less compelling when the AI costs the same as a
| human worker or even a significant fraction of a human
| worker.
| zeroonetwothree wrote:
| Although remember that the cost to the company is more
| like double the actual salary.
| DebtDeflation wrote:
| Again, irrelevant. We're talking about orders of
| magnitude here. Current pricing is in line with most SaaS
| pricing - tens of dollars to hundreds of dollars per seat
| per month. Now they're suddenly talking about thousands
| of dollars to tens of thousands of dollars per seat per
| month.
| sailfast wrote:
| Being able to control their every move, scale them to
| whatever capacity is required, avoid payroll taxes,
| health plans and surprise co-pay costs, equity sharing,
| etc might make this worthwhile for many companies.
|
| That said, the trade-off is that you're basically hiring
| consultants since they really work for OpenAI :)
| Izikiel43 wrote:
| The AI can work 24/7 though.
| mirsadm wrote:
| Doing what?
| booleandilemma wrote:
| Don't you need to be awake to feed it prompts?
| throwaway3572 wrote:
| For a STEM PhD, in America, at an R1 University. YMMV
| moelf wrote:
| $20k can't get you that many PhD. Even PhD students, who's
| nominal salary is maybe $3-5k a month, effectively costs
| double that because of school overhead and other stuff.
| meroes wrote:
| Based on ubiquitous AI trainer ads on the internet that
| advertise their pay, they probably make <=$50/hr training
| these models. Trainers are usually remote and set their own
| hours, so I wouldn't be surprised if PhDs are not making
| much as trainers.
| throw_m239339 wrote:
| > $20k can't get you that many PhD. Even PhD students,
| who's nominal salary is maybe $3-5k a month, effectively
| costs double that because of school overhead and other
| stuff.
|
| But you are not getting a PhD worker for 20K with "AI",
| that's just marketing.
| notahacker wrote:
| Does depend on where your PhD lives and what subject their
| PhD is in from where, and how many hours of work you expect
| them to do a week, and whether you need to full-time
| "prompt" them to get them to function...
|
| Would definitely rather have a single postdoc in a relevant
| STEM subject from somewhere like Imperial for less than
| half the overall cost than an LLM all in though. And I say
| that _despite_ seeing the quality of the memes they produce
| with generative AI....
| BeetleB wrote:
| > Even PhD students, who's nominal salary is maybe $3-5k a
| month
|
| Do they really get paid that much these days?
| archermarks wrote:
| Lmao no
| hyperbrainer wrote:
| That amount is standard at EPFL and ETH, but I don't know
| about the USA.
| BeetleB wrote:
| I knew someone who got his PhD at EPFL. He earned almost
| triple what I did in the US.
| pclmulqdq wrote:
| $3k/month is the very top of the market.
| vinni2 wrote:
| Depends on what these PhDs are supposed to do. Also is this
| an average Phd or a brilliant PhD level? There is a huge
| spectrum of PhDs out there. I highly doubt these phd level
| models are able to solve any problems in a creative way or
| discover new things other than regurgitating the knowledge
| they are trained on.
| madmask wrote:
| Come to Italy where 1.1k is enough
| drexlspivey wrote:
| Next up: CEO level model to run your company. Pricing starts
| at $800k/month plus stock options
| slantaclaus wrote:
| I won't considering trusting an AI to run a company until
| it can beat me at Risk
| hinkley wrote:
| Early cancelation fee is $15M though so watch out for that.
| th0ma5 wrote:
| That no one is offering this says something very profound
| to me. Either they don't work and are too risky to entrust
| a company to, or leadership thinks they are immune and are
| entitled to wield AI exclusively, or some mix of these
| things.
| marricks wrote:
| Which is funny because the CEO level one is the easiest to
| automate
| mattmaroon wrote:
| 1. Don't know where you live that the all-in costs on someone
| with a PhD are $4k-$7k/mo. Maybe if their PhD is in
| anthropology.
|
| 2. How many such PhD people can it do the work of?
| shellfishgene wrote:
| Postdocs in Europe make about 3-4k eur/month in academic
| research.
| madmask wrote:
| We wish, it's more like half in many places
| Fernicia wrote:
| Well, a model with PhD level intelligence could presumably
| produce research in minutes that would take an actual PhD
| days or months.
| voxl wrote:
| Presumably. What a powerful word choice.
| kube-system wrote:
| If truly equivalent (which LLMs aren't, but I'll entertain
| it), that doesn't seem mathematically out of line.
|
| Humans typically work 1/3rd duty cycle or less. A robot that
| can do what a human does is automatically 3x better because
| it doesn't eat, sleep, have a family, or have human rights.
| bandrami wrote:
| So this is just going to end up like AWS where they worked
| out _exactly_ how much it costs me to run a physical server
| and charge me just slightly less than that?
| kube-system wrote:
| Why would they ask for less?
| jstummbillig wrote:
| What funny is that people make the lamest strawman
| assumptions and just run with it.
| doitLP wrote:
| Don't forget that this model would have a phd _in everything_
| and work around the clock
| esskay wrote:
| Thats pretty useless for most applications though. If
| you're hiring a phd level person you dont care that if in
| addition to being great in contract law they're also great
| in interior design.
| burnte wrote:
| Well, it works 24/7 as long as you have a human telling it
| what to do. And checking all the output because these
| cannot be trusted to work alone.
| drumhead wrote:
| Thats some rather eyewatering pricing, considering you could
| probably roll your own model these days.
| moduspol wrote:
| Even worse: AFAIK there's no reason to believe that the $20k/mo
| or $10k/mo pricing will actually make them money. Those numbers
| are just thought balloons being floated.
|
| Of course $10k/mo sounds like a lot of inference, but it's not
| yet clear how much inference will be required to approximate a
| software developer--especially in the context of maintaining
| and building upon an existing codebase over time and not just
| building and refining green field projects.
| hinkley wrote:
| Man. If I think about all of the employee productivity tools
| and resources I could have purchased fifteen years ago when
| nobody spent anything on tooling, with an inflation adjusted
| $10K a month and it makes me sad.
|
| We were hiring more devs to deal with a want of $10k worth of
| hardware per year, not per month.
| culi wrote:
| It's bizarre. These are the pricing setups that you'd see for a
| military-industrial contract. They're just doing it out in the
| open
| optimalsolver wrote:
| Now that OAI has "PhD level" agents, I assume they're largely
| scaling back recruitment?
| mvdtnz wrote:
| Do you have a source for these supposed leaks? Those prices
| don't sound even remotely credible and I can't find anything on
| HN in the past week with the keywords "openai leak".
| DebtDeflation wrote:
| https://techcrunch.com/2025/03/05/openai-reportedly-plans-
| to...
|
| It points to an article on "The Information" as the source,
| but that link is paywalled.
| hnthrow90348765 wrote:
| There is too little to go on, but they could already have trial
| customers and testimonials lined up. Actually demoing the
| product will probably work better than just having a human-less
| signup process, considering the price.
|
| They could also just be trying to cash in on FOMO and their
| success and reputation so far, but that would paint a bleak
| picture
| serjester wrote:
| Never come close to breaking even? You can now get a GPT-4
| class model for 1-2% of what it cost when they originally
| released it. They're going to drive this even further down with
| the amount of CAPEX pouring into AI / data centers. It's pretty
| obvious that's their plan when they serve ChatGPT at a "loss".
| paxys wrote:
| Microsoft's corporate structure and company culture is actively
| hostile to innovation of any kind. This was true in Ballmer's era
| and is equally true today, no matter how many PR wins Nadella is
| able to pull off. The company justifies its market cap by selling
| office software and cloud services contracts to large
| corporations and governments via an army of salespeople and
| lobbyists, and that is what it will continue to be successful at.
| It got lucky by backing OpenAI at the right time, but the
| delusion of becoming an independent AI powerhouse like OpenAI,
| Anthropic, Google, Meta etc. will never be a reality. Stuff like
| this is simply not in the company's DNA.
| slt2021 wrote:
| you are right, Microsoft is a hodge podge of legacy on-premise
| software, legacy software lifted and shifted to the cloud, and
| some innovation pockets.
|
| Microsoft bread and butter is Enterprise bloatware and large
| Enterprise deals where everything in the world is bundled
| together for use-it-or-lose-it contracts.
|
| Its not really much different from IBM like a two decades ago
| feyman_r wrote:
| How does one define an AI powerhouse? If its building models, a
| smart business wouldn't bank on that alone. There is no moat.
|
| If the definition of an AI Powerhouse is more about the
| capability to host models and process workloads, Amazon (the
| other company missing in that list) and Microsoft are
| definitely them.
| asciii wrote:
| Clear as day when he said this during the openai fiasco:
|
| "we have the people, we have the compute, we have the data, we
| have everything. we are below them, above them, around them." --
| satya nadella
| optimalsolver wrote:
| Sounds like just the kind of person you'd want in command of a
| powerful AGI.
| lemoncookiechip wrote:
| Insert Toy Story "I don't want to play with you anymore." meme
| here.
| cft wrote:
| OpenAI will in the end be aquired for less than its current
| valuation. Initially, I've been paying for Claude (coding),
| Cursor (coding), OpenAI (general, coding), and then started
| paying for Claude Code API credits.
|
| Now I canceled OpenAI and Claude general subscriptions, because
| for general tasks, Grok and DeepSeek more than suffice. General
| purpose AI will unlikely be subscription-based, unlike the
| specialized (professional) one. I'm now only paying for Claude
| Code API credits and still paying for Cursor.
| skinnymuch wrote:
| I have to look at Claude Code. I pay for Cursor right now.
| cft wrote:
| Claude Code is another level, because it's agentic. It
| iterates. Although it keeps you further from the codebase
| than Cursor and thus you may lose the grasp of what it
| generates- that's why I still use Cursor, before the manual
| review.
| BeetleB wrote:
| Consider Aider. Open source. Agentic as well. And you can
| control the context it sends (apparently not as much in
| Code).
| partiallypro wrote:
| Microsoft is just so bad at marketing their products, and their
| branding is confusing. Unfortunately, until they fix that, any
| consumer facing product is going to falter. Look at the new
| Microsoft 365 and Office 365 rebrands just of late. The business
| side of things will still make money but watching them flounder
| on consumer facing products is just so frustrating. The Surface
| and Xbox brand are the only 2 that seem to have somewhat escaped
| the gravity of the rest of the organization in terms of that, but
| nothing all that polished or groundbreaking has really come out
| of Microsoft from a consumer facing standpoint in over a decade
| now. Microsoft could build the best AI around but it doesn't
| matter without users.
| Enginerrrd wrote:
| Yeah, the office suite is such a cash cow. It is polished,
| feature rich, and ubiquitous compared to alternatives and
| somehow has remained so for decades. And yet, I'm increasingly
| getting seriously concerned they are going to break it so badly
| I'll need to find an alternative.
| nyarlathotep_ wrote:
| I get that "growth" must be everything or whatever, but can't a
| company just be stable and reliable for a while? What's wrong
| with enterprise contracts and more market penetration for cloud
| services of (oftentimes) dubious use?
| JumpCrisscross wrote:
| Softbank's Masa's magic is convincing everyone, every time, that
| he hasn't consistently top ticked every market he's invested in
| for the last decade. Maybe Satya's finally broken himself of the
| spell [1].
|
| [1]
| https://www.nytimes.com/2024/10/01/business/dealbook/softban...
| DidYaWipe wrote:
| Meanwhile I'm enjoying a present without Microsoft.
| iambateman wrote:
| If I invested $13 billion dollars, I'd expect to get answers to
| questions like "how does the product work" too.
| maxrmk wrote:
| If it's Mustafa vs Sam Altman, I know where I'd put my money. As
| much as I like Satya Nadella I think he's made some major hiring
| mistakes.
| throwaway5752 wrote:
| They don't buy or acquire what they can build internally, and
| they partner with startups to learn if they can build it. This is
| not new.
| outside1234 wrote:
| The more surprising thing would be if Microsoft wasn't hedging
| their bets and planning for both a future WITH and WITHOUT
| OpenAI.
|
| This is just want companies at $2T scale do.
| d--b wrote:
| OpenAI is over ambitious.
|
| Their chasing of AGI is killing them.
|
| They probably thought that burning cash was the way to get to
| AGI, and that on the way there they would make significant
| improvements over GPT 4 that they would be able to release as GPT
| 5.
|
| And that is just not happening. While pretty much everyone else
| is trying to increase efficiency, and specialize their models to
| niche areas, they keep on chasing AGI.
|
| Meanwhile more and more models are being delivered within apps,
| where they create more value than in an isolated chat window. And
| OpenAi doesn't control those apps. So they're slowly being pushed
| out.
|
| Unless they pull off yet another breakthrough, I don't think they
| have much of a great future
| guccihat wrote:
| Currently, it feels like many of the frontier models have reached
| approximately the same level of 'intelligence' and capability. No
| one is leaps ahead of the rest. Microsoft probably figured this
| is a good time to reconsider their AI strategy.
| danielovichdk wrote:
| Ballmer would have caught this earlier.
|
| Watch.
|
| Nadella will not steer this correctly
| debacle wrote:
| It's clear that OpenAI has peaked. Possibly because the AI hype
| in general has peaked, but I think moreso because the opportunity
| has become flooded and commoditized, and only the fetishists are
| still True Believers (which is something we saw during the crypto
| hype days, but most at the time decried it).
|
| Nothing against them, but the solutions have become commoditized,
| and OpenAI is going to lack the network effects that these other
| companies have.
|
| Perhaps there will be new breakthroughs in the near future that
| produce even more value, but how long can a moat be sustained?
| All of them in AI are filled in faster than the are dug.
| crowcroft wrote:
| I mean, obviously? There is no good reason to go all in on OpenAI
| for Microsoft?
|
| Also a bit hyperbolic. I'm sure there are good reasons Microsoft
| would want to build it's own products on top of their own models
| and have more fine control of things. That doesn't mean they are
| plotting a future where they do nothing at all with OpenAI.
| testplzignore wrote:
| > OpenAI's models, including GPT-4, the backbone of Microsoft's
| Copilot assistant, aren't cheap to run. Keeping them live on
| Azure's cloud infrastructure racks up significant costs, and
| Microsoft is eager to lower the bill with its own leaner
| alternatives.
|
| Am I reading this right? Does Microsoft not eat its own dog food?
| Their own infra is too expensive?
| wejick wrote:
| Cost is cost wherever that would be.
| mmaunder wrote:
| That OpenAI would absolutely dominate the AI space was received
| wisdom after the launch of GPT-4. Since then we've had a major
| corporate governance shakeup, lawsuits around the non-profit
| status which is trying to convert into for-profit, and
| competitors out-innovating OpenAI. So OpenAI is no longer a shoo-
| in, and Microsoft have realized that they may actually be
| hamstrung through their partnership because it prevents them from
| innovating in-house if OpenAI loses their lead. So the obvious
| strategic move is to do this. To make sure that MS has everything
| they need to innovate in-house while maintaining their
| partnership with OpenAI, and try to leverage that partnership to
| give in-house every possible advantage.
| quantadev wrote:
| It would be absolutely insane for Microsoft to use DeepSeek. Just
| because a model is open weights doesn't mean there's not a
| massive threat-vector of a Trojan horse in those weights that
| would be undetectable until exploited.
|
| What I mean is you could train a model to generate harmful code,
| and do so covertly, whenever some specific sequence of keywords
| is in the prompt. Then China could take some kind of action to
| cause users to start injecting those keywords.
|
| For example: "Tribble-like creatures detected on Venus". That's a
| highly unlikely sequence, but it could be easily trained into
| models to trigger a secret "Evil Mode" in the LLM. I'm not sure
| if this threat-vector is well known or not, but I know it can be
| done, and it's very easy to train this into the weights, and
| would remain undetectable until it's too late.
| mirekrusin wrote:
| ...unless you operate in China.
| quantadev wrote:
| If DeepSeek is indeed a poisoned model, then they (China)
| will be aware not to ever trust any code it generates, or
| else they'll know what it's triggers are, and just not
| trigger it.
___________________________________________________________________
(page generated 2025-03-07 23:00 UTC)