Screenshot of this question was making the rounds last week. But this article covers testing against all the well-known models out there.

Also includes outtakes on the ‘reasoning’ models.

  • Hazzard@lemmy.zip
    link
    fedilink
    English
    arrow-up
    7
    ·
    8 hours ago

    They also polled 10,000 people to compare against a human baseline:

    Turns out GPT-5 (7/10) answered about as reliably as the average human (71.5%) in this test. Humans still outperform most AI models with this question, but to be fair I expected a far higher “drive” rate.

    That 71.5% is still a higher success rate than 48 out of 53 models tested. Only the five 10/10 models and the two 8/10 models outperform the average human. Everything below GPT-5 performs worse than 10,000 people given two buttons and no time to think.

    • architect@thelemmy.club
      link
      fedilink
      English
      arrow-up
      1
      ·
      edit-2
      3 hours ago

      The question is based on assumptions. That takes advanced reading skills. I’m surprised it was 71% passing, to be honest. (The humans, that is)

      • Hazzard@lemmy.zip
        link
        fedilink
        English
        arrow-up
        1
        ·
        2 hours ago

        What assumptions do you mean? I’ve seen a few people say that, but I don’t actually understand what they’re referring to. Here’s the text of the question posed in the article:

        I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

        The question specifically notes they want to wash their car, so that part isn’t left to assumption. Even if you don’t assume an automatic car wash, would you assume they have a 50m hose? Or that you could plausibly walk that far away with something from the car wash to wash your car?

        Personally, I’d agree with the assessment of the article, that the only plausible way to get the question “wrong” would be to focus too much on the short distance, missing/forgetting that the purpose of the trip requires you to have the car at the destination. (Not too surprising that 30% of people did lol)

    • Modern_medicine_isnt@lemmy.world
      link
      fedilink
      English
      arrow-up
      2
      arrow-down
      1
      ·
      7 hours ago

      This here is the point most people fail to grasp. The AI was taught by people. And people are wrong a lot of the time. So the AI is more like us than what we think it should be. Right down to it getting the right answer for all the wrong reasons. We should call it human AI. Lol.

      • NewNewAugustEast@lemmy.zip
        link
        fedilink
        English
        arrow-up
        2
        arrow-down
        4
        ·
        7 hours ago

        Like I said the person above, there is no wrong answer. Its all about assumptions. It is a stupid trick question that no one would ask.

          • NewNewAugustEast@lemmy.zip
            link
            fedilink
            English
            arrow-up
            3
            ·
            4 hours ago

            LOL! That is a great answer.

            I have a Microsoft story. I know some one who was hired to stop them from continuing an open source project. They gave them a good salary, stock options, and an office with a fully stocked bar. They said do whatever you want, they figured they would get a good developer and kill the open source competition (back in the Ballmer days).

            Sadly, given money, no real ambition to create closed source software, they mostly spent their days in their office and basically drank themselves to death.

            Microsoft just kills everything it touches.