• Captchas : Now SO Bad ONLY "AI" Can Solve Them

    From 26xh.0712@26xh.0713@e6t5y.net to talk.politics.misc,alt.politics,alt.security,alt.politics.usa,alt.elections on Friday, July 05, 2024 22:10:55
    From Newsgroup: alt.security

    https://www.newsnationnow.comtnewsnation-now/captchas-are-becoming-more-difficult-to-solve/?ipid=promo-link-block1

    “Captchas are old enough, though, that their age now is
    presenting a tremendous number of weaknesses,” said Sultan
    Meghji, CEO of Frontier Foundry. “Especially in the current
    generation of generative AI. These new generative AI systems,
    like OpenAI, are designed to create medium to low-quality,
    or in some cases, very high-quality, audio, video, text and
    pictures. And it perfectly aligns with the technology
    behind captchas.”

    This arms race of sorts has reached a point where computers
    are far better at solving them than humans are.

    . . .

    Captchas are getting SO hard to solve - and I'm over 65 -
    that I've considered organizing a class-action ADA suit.

    In truth, these days, anything that CAN solve a captcha
    is probably NOT a human.

    Current "AI" has been *extensively* trained on "people
    stuff" - a zillion terabytes pirated from the Net and
    social media sites (usually without permission/pay).
    Chat4 now passes the "Turing Test" - ie most people
    can't TELL it's not a human. Maybe that qualifies it
    as a form of intelligent life ? Lawyers ?

    In any case, the "I am not a robot" tests WON'T BE
    RELIABLE anymore. Beyond captchas ... well ... there
    really may be NO good test anymore. The "AI" can
    emulate your IQ, even your face WITH voice. It's
    a serious problem both for biz AND individuals.
    Might even be able to fake biometric data provided
    from a cell or whatever.

    This gives evil actors tremendous power. Banks and
    such may want customers to provide a list of code-words
    or phrases only they and the bank know. Don't lose
    yer copy ! Online banking may have to be banned, only
    in-person flesh-n-blood contact where they can poke
    you to prove you're not a robot.

    OR, we can make such "AI" illegal - shut it down
    forever. Small actors are NOT using their own AI
    software - it's WAY too big - they are using what
    people like M$/Amazon/META are offering. Alas probably
    a trillion bucks have been invested already ......

    We THINK that Russian nukes are THE problem ... but
    doom is much more LIKELY to come from abuses of "AI".
    --- Synchronet 3.21e-Win32 NewsLink 1.2