Tausende CEOs geben zu, dass KI keinen Einfluss auf Beschäftigung oder Produktivität hatte – und Ökonomen lassen damit ein Paradoxon von vor 40 Jahren wieder aufleben

    https://fortune.com/article/why-do-thousands-of-ceos-believe-ai-not-having-impact-productivity-employment-study/

    Share.

    32 Kommentare

    1. Thousands of CEOs admit they don’t know how to use AI correctly or competently. 

    2. BenefitPlastic5609 on

      The Solow Paradox all over again — we spent billions on computers in the 80s with nothing to show for it either, turns out just buying shiny new tech without rethinking workflows never works.

    3. Wischiwaschbaer on

      Couldn’t be that AI it’s just shit at this point in time. Has to be some paradoxon.

    4. stuartullman on

      what is the point of linking to a paywall, so no one can read the article and get the actual context behind the clickbait article?

    5. CanIhazCooKIenOw on

      My FAANG adjacent company had the same realisation, that all the AI usage as not translated to any productivity improvement.

      But they are doubling down and the pipe dream is that by October no code is written by humans…

    6. But you could use it for the excuse for firing people who weren’t needed to begin with.

    7. The only epiphany I had from reading this article is that 1987 was almost 40 years ago.

    8. the_ballmer_peak on

      I always think back to the same phenomenon: you can improve your organization across almost every level of delivery, but if there is a bottleneck somewhere and you don’t improve that bottleneck, nothing will get better.

      My organization keeps trying to improve on project execution even though I’ve told them a hundred times that their bottleneck is decision making.

    9. StretchMother9627 on

      Oh good well at least we went about implementing it cautiously ethically and judiciously , great job deserved winners of the only system possible , back to the scamming board I guess

    10. IntelArtiGen on

      There’s a difference between the productivity we observe for us, and the real economic gains we can see on the GDP. For example AI won’t build more houses, AI won’t produce more energy, or cement, or trees, or fish, etc. And simultaneously even if I’m more efficient with AI, all my competitors are also more efficient, so I don’t have an advantage.

      Recent events also clearly show that no matter what we do with AI, oil and gas are much more important for the global economy. ChatGPT won’t stock the grocery store or build houses, it’s a gadget.

    11. Thank you for your article. It will be responded to by the next available chatbot.

    12. Bad-job-dad on

      Probably because chatgpt is too busy telling everyone how awesome they are.

    13. ConiferousTurtle on

      I feel like everyone wants AI now, but they have no idea how it can help them or why they need it. Bubble bursting soon?

    14. farm_shapes on

      my company is tracking each individual’s AI output on a company leaderboard where anyone can check it

    15. TheorySudden5996 on

      I just built a major application for a Fortune 50 using largely AI in under 2 months with a team of one. I know how to code, I could have built it all by hand. It would have taken at least 2x as long and would have likely required a larger team.

      Most companies just haven’t figured where and how to leverage AI yet.

    16. DreadPirate777 on

      Breaking news! CEO lies to keep stock price up. In other news water is wet.

    17. This isn’t about productivity gains. It’s about replacing people with machines.

    18. Discord_aut7 on

      My take – companies should have invested in the people overall. The health and wealth of the general population. Increasing healthcare, academics, the „smartness“ of a society and so on – rather than let the wealth divide increase quickly. The longterm payoff would have been larger societal gains and I think that’s what is missing in our race to whatever the fuck a lot of these technofascists think the future is.

    19. We know AI usage is dependant on availability of data and availability of compute (and context memory size). When AI has these two parameters ready, the entire sub sector is devint destroyed in a short time (coding. Customer service, content, marketing).

      But there are a lot of segments where data is either not available or not readily available (siloed database, on prem data etc) where it will take much longer.

      The crazy thing is that the already disrupted segments are c. 5 trillions annual spending.
      And that’s just the beginning

    20. nah_dude_lol on

      I get there’s tons of social currency in AI bashing but it has increased my productivity as a developer like 5 fold at least

    21. someaccountforreddit on

      AI has clear productivity gains on tasks like code development, review, test, debug etc. However there are many other bottlenecks in processes, people interactions etc which are still steeped in the pre-AI world. Until all these catch up we will see limited gains at best.

    22. Pluto-Had-It-Coming on

      AI absolutely helps with my productivity a bit. 

      Not a lot, but a bit. 

    Leave A Reply