Close Menu
    X (Twitter) LinkedIn
    CapitalAI DailyCapitalAI Daily
    X (Twitter) LinkedIn
    • Markets & Investments
    • Big Tech & AI
    • AI & Cybercrime
    • Jobs & AI
    • Banks
    • Crypto
    Sunday, November 2
    CapitalAI DailyCapitalAI Daily
    Home»Big Tech & AI»Meta Researcher Predicts Intelligence Costs Collapsing Toward Zero, Says Knowledge Will Be Instant

    Meta Researcher Predicts Intelligence Costs Collapsing Toward Zero, Says Knowledge Will Be Instant

    By Henry KanapiOctober 26, 20252 Mins Read
    Share
    Twitter LinkedIn

    A leading researcher at Meta’s Superintelligence Labs says technology is rapidly advancing to a level where the cost of intelligence is zero.

    Speaking at the Stanford AI Club, Jason Wei, one of the researchers behind large-scale reasoning work at Meta, says advances in adaptive compute are ushering in an era where intelligence becomes as accessible as electricity or data bandwidth.

    According to Wei, AI has surpassed the early frontier stage, or a phase when systems can’t perform tasks effectively. He says AI is now in the second phase, where AI gains an ability that can be commoditized.

    Using the Massive Multitask Language Understanding (MMLU) metric, the Meta researcher notes that using a model with a particular level of intelligence decreases every year.

    “Why might this trend continue? And the argument that I would make, um, is that it’s the first time in the history of deep learning that adaptive compute actually truly works. So if you look at maybe the entirety of deep learning up to last year, we were in sort of this mode where the amount of compute you would use to do a particular problem would be fixed regardless of how hard the problem was, whether it was giving the capital of California or a very hard competition math problem. But now we’re in this era of adaptive compute, where you can vary the amount of compute used to do your task.”

    To illustrate, Wei points to how the time needed to retrieve information has collapsed — from hours in the library era to minutes or seconds online and now to near-zero latency through AI interfaces.

    “So, for example, if you wanted to know, what was the population of Busan in 1983? Before the internet, you’d probably have to drive to the library and then look in a bunch of encyclopedias to find that… In the internet era you’d have to search, and then you kind of like browse the websites to find the one that actually gives you
    the answer, and that might take minutes. And then now it’s basically instant to get that answer.”

    Wei predicts that the trend will continue to the point when knowledge becomes available in a flash.

    “The idea of instant knowledge, so anything that’s sort of publicly available information, you’ll be able to get access to that instantly.”

    Disclaimer: Opinions expressed at CapitalAI Daily are not investment advice. Investors should do their own due diligence before making any decisions involving securities, cryptocurrencies, or digital assets. Your transfers and trades are at your own risk, and any losses you may incur are your responsibility. CapitalAI Daily does not recommend the buying or selling of any assets, nor is CapitalAI Daily an investment advisor. See our Editorial Standards and Terms of Use.

    AI commodity intelligence Meta
    Previous ArticleSoftBank Greenlights $22.5 Billion OpenAI Funding, Contingent on Corporate Overhaul Before IPO
    Next Article Goldman Sachs Sees No AI Bubble, But Flags Two Warning Signs for Bull Market Top

    Read More

    Elon Musk Says AI and Robotics Can Save US From $38.101 Trillion National Debt – Unless America Goes ‘Full Genghis Khan’

    November 2, 2025

    Mark Cuban Warns China Could Win AI War and Deploy Trojan-Horse Models To Influence Americans

    November 2, 2025

    Sam Altman Fires Back at OpenAI Doubters, Says Short OpenAI and ‘Get Burned’

    November 2, 2025

    Goldman Sachs Says AI Set To Deliver 15% Productivity Boom, Fuel Multi-Trillion-Dollar Economic Expansion

    November 1, 2025

    Fundstrat’s Tom Lee Sees Path to S&P 500 All-Time Highs by Year-End, Cites AI Margins and Two More Tailwinds

    November 1, 2025

    Elon Musk Predicts End of Smartphones, Apps With AI Taking Over Devices – Here’s His Timeline

    November 1, 2025
    X (Twitter) LinkedIn
    • About
    • Author
    • Editorial Standards
    • Contact Us
    • Privacy Policy
    • Terms of Service
    • Cookie Policy
    © 2025 CapitalAI Daily. All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.