Business leaders shouldn't "BS" employees about the impact of AI on jobs, according to one tech billionaire, who says they should be transparent and honest.

  • Jim Kavanaugh, CEO of World Wide Technology, told CNBC that people are “too smart” to accept artificial intelligence won’t alter their work environment.

  • Business leaders shouldn’t “BS” employees about the impact of AI on jobs, Kavanaugh said, adding that they should be as transparent and honest as possible.

  • Kavanaugh, who has a net worth of $7 billion, stressed that overall he’s an optimist when it comes to AI and its ability to improve productivity.

  • 🇨🇦🇩🇪🇨🇳张殿李🇨🇳🇩🇪🇨🇦@ttrpg.network
    3·
    2 months ago

    They didn’t oppose technologies. They destroyed the visible manifestations of employee abuse which happened to be technological (stocking frames and threshing machines). They were labour activists who did “collective bargaining by riot” since there was no legal framework for collective bargaining at the time (and indeed the law was often against any efforts of the working class to organize and improve their lot).

    Their particular cause failed (c.f. law being onside with the abusers), but they laid the groundwork for the future of labour.

    So this brings us to modern times. Nobody (sane) is against “AI”. (“Scare quotes” because we don’t have AI yet and likely never will in my time nor in the next two generations’.) They’re opposed to so-called LLMs, one specific brand of “AI” (a.k.a. Generative AI or as I prefer to call it Degenerative AI). And the reasons to oppose it are multifold.

    1. They’re predicated on mass theft.
    2. They’re predicated on mass lies about their efficacy.
    3. They’re being inserted into everything whether we want it or not.
    4. The things they’re being inserted into are rapaciously sucking up everything we do and say for MOAR KONTTENT!
    5. They’re killing the planet. (A recent study determined that a 100-word output from ChatGPT consumes half a litre of water, in addition to its massive energy cost.)

    And the thing is, they can’t be meaningfully improved upon. Each rendition of ChatGPT, for example (which, incidentally, is a near-homophone for the French “chat, j’ai pété” which I will never unhear), has diminishing returns on the quality of its output. ChatGPT 1 to 2 was a huge leap. 2 to 3 was a smaller leap. 3 to 4 is almost indistinguishable for casual use. 4 to 4o<insert alphanumerical soup here> even less improved.

    All while the computational power to run each increasing rendition goes up exponentially.

    So LLMs in particular are technology that needs to be killed, now. Not in some Butlerian Jihad, but in terms of just fining the motherfuckers foisting it upon us and cramming it into every orifice they can find with fines well in excess of 100% of gross revenue.