Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
Facebook AI Businesses Slashdot.org

Meta Freezes AI Hiring 53

Posted by BeauHD from the bubble-about-to-burst dept.
According to the Wall Street Journal, Meta has paused hiring in its artificial intelligence division after bringing on more than 50 researchers and engineers. "All that's happening here is some basic organizational planning: creating a solid structure for our new superintelligence efforts after bringing people on board and undertaking yearly budgeting and planning exercises," a spokesperson for Meta said in an emailed statement to Reuters.

Over the last few months, Meta has been offering AI researchers salaries that dwarf those of the Manhattan Project and the Space Race. The company recently offered AI researcher Matt Deitke $250 million over four years (an average of $62.5 million per year), with potentially $100 million in the first year alone. Meta CEO Mark Zuckerberg reportedly also offered an unnamed AI engineer $1 billion in compensation to be paid out over several years.

Meta Freezes AI Hiring More | Reply

Meta Freezes AI Hiring

Comments Filter:

  • Maybe the previous generation has developed the self-improving AI and we're already past the singularity.

    Or we're close to the crash of this unbelievably stupid bubble, and money dumped into it with massive negative returns is running out so that even dumb fucks like zuck and sam need a reality check.

    • Sadly, we're not even close

    • Re:So what? (Score:5, Insightful)

      by sg_oneill ( 159032 ) on Thursday August 21, 2025 @02:55AM (#65604246)

      I still think the next major stock market crash is going to come from Tesla, rather than AI. A company that has maybe 2-3% of the car market with a higher market cap (roughly amount of shares multiplied by price of those shares) than the rest of the automobile industry combined is a market absurdity that is surely going to correct to, at a guess, maybe 3% of its current stock price, because inevitably it has to. At some point the shares must reflect reality or it implodes with stupid. And whether it corrects first, or it just implodes, a hell of a lot of very rich people are suddenly going to become a lot less rich, and thats how market panics start, and market panics are how stock markets crash.

      • Re:So what? (Score:4, Insightful)

        by Mr. Dollar Ton ( 5495648 ) on Thursday August 21, 2025 @03:20AM (#65604276)

        We'll see, but a large piece of the Tesla stock valuation is based on the bullshit that "it is not a car company, it is an 'AI'-robot magic".

        So if it crashes first, it will also be a precursor or a part of the larger AI bubble collapse, because their abysmal car business performance is hardly an important part of the stock price.

      • Market share is a poor predictor of whether a company is on sure footing. In Tesla's case, that's especially true when you consider that they still hold a high market share of EVs, and slightly less-so among luxury cars, where margins are generally higher, and market share doesn't mean a lot (case in point: Lamborghini.) If you want to look at only one number when you're trying to figure out whether a stock is overvalued, then a much better bet than market share is P/E ratio. But even then, just looking at

    • Or we're close to the crash of this unbelievably stupid bubble, and money dumped into it with massive negative returns

      ChatGPT has over 400 million active users [reddit.com]. If you have 400 million active users, it's not a bubble.

      Eventually it will enter the monetization/enshittification stage, and they will monetize those users.

      • Re:So what? (Score:4, Insightful)

        by Mr. Dollar Ton ( 5495648 ) on Thursday August 21, 2025 @04:40AM (#65604354)

        The ratio of "active users" to all internet users for myspace at its peak around 2007 was what, 120M/1.6B? Something similar, or about 7.5% of the population. At the time, the users were a lot less mobile (it was most of us when we were a lot younger, and we still can't forget slashdot).

        The chatgpt has a little lower ratio of users as a percent of the internet users today (400M/6B, or as they say, about 5%), but today users are changing habits much faster, so they jump on faster, and drop you faster.

        No doubt the free part will linger on for ages, but it is very hard to say anything at all about the "value" of those "400 million active weekly users" or of the business as a going concern.

  • Imagine (Score:4, Insightful)

    by zamboni1138 ( 308944 ) on Wednesday August 20, 2025 @11:02PM (#65604054)

    Imagine the horror of attempting to try to attain super-intelligence on data gleamed from Meta. But I guess when you have hundreds of billions of dollars to waste, the worst that can happen is, I don't know, depicted in some movie somewhere. The real story here: We are not taxing the rich enough. Time to go back to 70% for the upper tax bracket. You can still be stupid rich, but we need to feed and house the poor, and rebuild our roads, and provide affordable electricity/water/sewer, and clean up our environmental disasters from WWII and maybe even some education and medical stuff in there too.

    • Re: (Score:2, Insightful)

      by ndsurvivor ( 891239 )
      I agree, Unfettered Capitalism has once again failed. The massive population who worship the Uber Rich need to realize that we are a Civilization, and that every person has a potential that needs to be nurtured. We need to build roads and schools, instead of nurturing the wet dreams of some out of control Capitalist who throws money around like the Kings and Oligarchs of history. We need a 70% tax on these people and a reset of what we consider: "useless eaters". Maybe anybody who does nothing more b
      • If you look at the top 10%, or the top 1%, they pay most of the taxes in this country. And in my town, 87% of the property taxes go to the school. Its hard to imagine spending more via taxes than we already do. If you don't see the results, its because the money is wasted, not because its not spent. They spend a lot on the roads, but here they last a few years. Not much incentive for the paving company to build a road that lasts 50 years because then they won't get another contract. Where I live there's onl

    • Re: (Score:2)

      by ledow ( 319597 )

      The way to tax is not to just set a tax bracket. That much is necessary, sure.

      But the way to actually stop them avoiding that tax, like they always do to huge extents, is to implement a simple system.

      Pick a tax rate.
      Now if you cannot prove that - in total - you have PAID that much tax direct to your relevant authority, by whatever means, under whatever tax... then you're charged the difference.

      So if you are supposed to pay 50% of your income but actually because of various schemes, stocks, investments, don

      • That's circular reasoning. What you are legally "supposed" to pay is defined by nothing but the tax code which has provisions for what you call the various schemes.

        • Re: (Score:2)

          by ledow ( 319597 )

          The tax code only applies if you're doing certain things in certain ways.

          If you've managed to go through a year without paying the minimum percent of tax, through whatever schemes, loopholes, avoidances or otherwise, than a mandatory percent of your income to which you have to top-up means your tax code doesn't HAVE to be watertight, and can't be subverted by those with other interests.

          It's a question of whether a thousand independent lawyers working for government tax department can draft new taxes for sch

      • You pretty much are describing the Alternative Minimum Tax [wikipedia.org], which already exists.

    • They need to reverse the order. Right now we're using data to compress knowledge and then build "reasoning" on top. System is static, doesn't learn anything new and reasoning is questionable.
      While in reality it's exactly the opposite, we create intelligence and then we teach it. This way system is dynamic, can learn, form opinions and draw conclusions.
      Until we switch the order around, we're not closer to any kind of AGI. LLMs are a dead end in my opinion.
    • I think you misunderstand how much all the magical projects you suggest would cost, and how taxing the rich even more is not even close to enough money. All that will do is stifle innovation and growth- two things that solve problems.
      How about we make it easy to build small nuclear reactors instead of wasting our time blaming the wealthy?
      • You forget that too much free-floating capital stifles economic growth. Investors and entrepreneurs get sloppy. Too many un-supportable/unproductive companies take up resources including IP that would do more if better targeted. Financials run on The PETER Principle ... where every fool has a chance ... do not produce maximum value. Creating value is extraordinarily hard , and few can perform the task. That's why a well-managed financial culture produces so few wealthy people.
    • âoeThe problem with socialism is that you always run out of other peopleâ(TM)s moneyâ - Margaret Thatcher âoeTax the richâ is a great idea except that thereâ(TM)s not enough of The Rich to support all the people with their hands out. (Or more accurately, pols trying to hand out money to registered voters.)

  • Oh, the hype! (Score:5, Insightful)

    by Sebby ( 238625 ) on Wednesday August 20, 2025 @11:10PM (#65604060) Journal

    The company recently offered AI researcher Matt Deitke $250 million over four years (an average of $62.5 million per year), with potentially $100 million in the first year alone. Meta CEO Mark Zuckerberg reportedly also offered an unnamed AI engineer $1 billion in compensation to be paid out over several years.

    The absurdity of the salaries here is how you know the AI hype has begun to hit its peak.

  • Something stinks in Denmark (Score:5, Insightful)

    by rsilvergun ( 571051 ) on Wednesday August 20, 2025 @11:14PM (#65604066)
    So one of the most common tricks in technology is to monopolize the engineers so that you can keep them out of the hands of potential competitors especially startups.

    I don't think with our current administration we will find anything out but assuming we ever have another proper administration I suspected a few years we will find out that the big players in this space are colluding to lower wages and pay.

    There is no way in hell that these crazy bidding wars were going to be allowed to continue

  • So now Facebook is going to start telling me what VCR tapes to rent?

  • Ray (Score:1)

    by Anonymous Coward

    How will this affect my Ray-Ban spyware glasses that I deliberately chose to pay real money for?

    • They're going to show your eyes only what the zuck thinks you should see.

      A very useful feature.

      • I find myself wondering (on the face value concept that this might genuinely be the long term plan from the ad dept, on the premise it's what their competitors would try to do) if Zuck and Co. understand how this kind of thing doesn't scale.

        He's managed to fail at understanding it in the past... They are traditionally even worse at it...

        Like yeah I know you're joking, but... Honestly I wouldn't put it past them, and not for reasons of capability.

        • Like yeah I know you're joking, but... Honestly I wouldn't put it past them, and not for reasons of capability.

          I'm only half-joking. I hear from people on the inside of FB that zuck really believes the "AGI" is very close and that it is a winner-takes-all game, so he's betting the bank.

          If you assume the "AGI" that the zuck is imagining, making people to see only what they're shown would probably as easy as stealing a candy from a baby.

  • Manhattan Project and the Space Race (Score:4, Interesting)

    by excelsior_gr ( 969383 ) on Thursday August 21, 2025 @07:30AM (#65604498)
    In the case of the Manhattan Project the people working there were fighting a war. That was the sentiment and the motivation (source: R. Feynman). In the case of the Space Race, the war was a "cold" one: there was a lot of general public excitement abound to beat the Ruskies to the moon. When it comes to AI, a lot of people are afraid it will come for their jobs and most people treat it just like a new kind search engine. Very few think AI is a cause worth working for. And for what?! The purpose here is not to beat an aggressor. Beyond the use cases that only some geeks are excited about (protein folding, image recognition and such) the main motivation for these hires is to make more money for Meta et al. *Of course* the salaries will be higher than those at the Manhattan and the Apollo projects.

    • Re: (Score:2)

      by evanh ( 627108 )

      And the science of nuclear reactions was already known before the Manhattan Project was undertaken.

      AI science hasn't even begun yet. We're still at the alchemists level.

    • Re: (Score:2)

      by Megane ( 129182 )
      Also a lot of the people working on the Manhattan Project died from various radiation-induced cancers. If only a similar fate could even be a remote possibility for the purveyors of AI-slop.

  • thinking if only money can buy a preexisting product like insta for a couple of billions.
    money to waste, wastes money

  • Zuck knows Facebook is hitting a wall, and VR didn't pan out the way he hoped, so he has to spend all this money to try to cargo-cult a win. This won't really pan out either. He'd be better off if he just cashed-in and enjoyed his life on his island in Hawaii and left the rest of us alone.

  • Another rich target lost...

Slashdot Top Deals

The confusion of a staff member is measured by the length of his memos. -- New York Times, Jan. 20, 1981

Close