Close Menu
    SoftBank CEO says AI that is 10,000 times smarter than humans will come out in 10 years
    Futurology

    SoftBank CEO says AI that is 10,000 times smarter than humans will come out in 10 years

    Share.
    View 22 Comments

    22 Comments

    1. Maxie445 on

      Another guy with an AGI prediction, so what? Fair, but part of the reason why I find this interesting is:

      1) SoftBank is one of the largest conglomerates in the world

      2) It used to be only AI-industry people with 3-5 year AGI timelines. AGI wasn’t really on anyone’s radar until recently.

      3) He said he expects AGI to arrive in 3-5 years, ASI within 10 years.

    3. SpillinThaTea on

      SoftBank also tossed a bunch of cash at WeWork and look how that turned out

    5. PeacefulGopher on

      lol they literally sold their NVidia positions before the run up and lost out on $150 Billion in upsides. Yeah, not the best to listen to about getting the future right…

    6. PornstarVirgin on

      Breaking ceo who has billions invested in the industry has something to say!

    7. ABrokenBinding on

      Current AI still won’t tell me who won the election in 2020. Perhaps he means smarter than some specific humans…?

    9. Barry_Bunghole_III on

      Whenever someone predicts the future, regardless of how qualified they are, turn your brain off immediately lol

    10. fremiamagus on

      We are still and far away from Generalized Artificial Intelligence. How far? A few hundred years based on similar breakthroughs in other industries.

      The largest 3 problems to fix right now are hallucinations, power use, and outright fabricating of data which will likely be solved in the next fifty.

      Next steps will be to monitor the growth of neural networks and cull errant dead ends in its’ development as well as increasing memory density(both short and long), and cooling, assuming no major tech breakthroughs on this front in the way to manage it without such. If photonics ever become a real thing it may be possible to not require cooling, but this won’t happen this century or next. By errant dead ends I mean the propensity of current algorithms to pick a certain path and just stick to it no matter how much of a headache it becomes.

      After that; you most likely have the groundwork necessary for a Real AGI/GAI.

    11. DriftMantis on

      This guy probably can’t even define what smarter means in any context because he’s just a talking head trying to advertise his speculative wares.

    12. sardoodledom_autism on

      That Ai will immediately ask the SoftBank ceo why he invested in WeWork and lost billions…

    13. WildBorneos on

      this CEO must be drinking something very woozy… in case m wrong us dumb humans make all this fake ai crap so the realization is NOT IN HIS LIFE CAN THIS HAPPEN

    14. Devilnaht on

      This is just asinine. Extraordinary claims require extraordinary evidence. Why is it that we forget that when it comes to this outlandish tech bullshit? How many cycles of people with clear financial incentives making absolutely buck wild claims about tech that is always “5-10 years away” have we been through in the last 10 or 15 years? And we’re supposed to take it seriously when the CEO of a fucking investment group when he says ‘found his life’s purpose to evolve humanity’? I’m sick of all the uninformed AI speculation and egregious scams and grift that it enables.

    16. ThresholdSeven on

      I can make up numbers too. Pretty sure nobody knows what AI will be in 10 years, but I can almost guarantee it won’t be even 1x as smart as a human. What’s the metric? Faster at math? Already done. Able to problem solve creatively? Maybe. Able to pass as human? Unlikely.

      “10,000” times “smarter” than humans is such a vague and arbitrary number that is just as ridiculous as anybody’s guess.

    17. BlackPresident on

      My calculator is already 10,000 times smarter than me at maths, I calculated it

    18. Ethereal_Bulwark on

      And it will be highly regulated, limited, and have multiple kill switches.

    19. shaz_y on

      Yeah, the CEO of Softbank really shouldn’t be making assumptions like these. Look at Softbank’s failed ventures; Wework & Sprint (which was dying before the T-Mobile acquisition).

    21. aconitine- on

      Hope this idiot and his crap company lose all their half-assed investments.

      Maybe an AI would have more sense than the jokers running the Softbank vision fund.