• @AdolfSchmitler
    link
    411 month ago

    There’s an idea about “autistic ai” or something where you give ai an objective like “get a person from point a to b as fast as you can” and the ai goes so fast the g force kills the person but the ai thinks it was a success because you never told it to keep the person alive.

    Though I suppose that’s more human error. Something we take as a given but a machine will not.

      • ඞmir
        link
        fedilink
        211 month ago

        That’s specifically LLMs. Image recognition like OP has nothing to do with language processing. Then there’s generative AI which needs some kind of mapping between prompts and weights, but is also a completely different type of “AI”

        That doesn’t mean any of these “AI” products can think, but don’t conflate LLMs and AI as being the same

          • ඞmir
            link
            fedilink
            101 month ago

            Neural networks aren’t going anywhere because they can be genuinely useful, just not to solve every problem

              • MeanEYE
                link
                21 month ago

                You should watch actually AI safety researcher’s thoughts on this. Here’s the link. It’s partially overhyped, but huge strides have been made in this area and it shouldn’t be taken lightly. It’s best to be extra careful than ignorant.

              • @FooBarrington
                link
                21 month ago

                And that somehow means we shouldn’t do OCR anymore, or image classification, or text to speech, or speech to text, or anomaly detection, or…?

                Neural networks are really good at pattern recognition, e.g. finding manufacturing defects in expensive products. Why throw all of this away?

        • @[email protected]
          link
          fedilink
          81 month ago

          Your brain is also “just a Chinese room”. It’s just physic, chemistry and biology. There is no magic inside your brain. If a “Chinese room” is fast enough and can fool everyone into “believing” that it’s fluent in chinese, than the room speaks chinese.

          • Kogasa
            link
            fedilink
            21 month ago

            This fails to engage with the thought experiment. The question isn’t if “the room is fluent in Chinese.” It is whether the machine learning model is actually comparable to the person in the room, executing program instructions to turn input into output without ever understanding anything about the input or output.

            • @[email protected]
              link
              fedilink
              2
              edit-2
              1 month ago

              The same is true for your brain. Show me the neurons that are fluent in Chinese. Of course the LLM is just executing code. And if we have AGI it will also just be “executing code” but so does your brain. It’s not exactly code (but maye AGI will be analog computers, so not exactly code either) but the laws of physics dictate what your brain does. The laws of physics don’t understand Chinese, the atoms and molecules don’t understand Chinese. “Understanding Chinese” is an emergent property.

              Think about it that way: Assume every person you know (execpt you) is just some form of Chinese Room … You first of all couldn’t prove that and second it wouldn’t matter at all.

              • Kogasa
                link
                fedilink
                01 month ago

                We aren’t trying to establish that neurons are conscious. The thought experiment presupposes that there is a consciousness, something capable of understanding, in the room. But there is no understanding because of the circumstances of the room. This demonstrates that the appearance of understanding cannot confirm the presence of understanding. The thought experiment can’t be formulated without a prior concept of what it means for a human consciousness to understand something, so I’m not sure it makes sense to say a human mind “is a Chinese room.” Anyway, the fact that a human mind can understand anything is established by completely different lines of thought.

      • @[email protected]
        link
        fedilink
        21 month ago

        How can you know the system has no cognitive capability ? We haven’t solved the problem for our own minds, we have no definition of what consciousness is. For all we know we might be a multimodal LLM ourselves.

      • @[email protected]
        link
        fedilink
        English
        -2
        edit-2
        1 month ago

        Language processing is a cognitive capability. You’re just saying it’s not AI because it isn’t as smart as HAL 9000 and Cortana. You’re getting your understanding of computer science from movies and video games.

    • @[email protected]
      link
      fedilink
      121 month ago

      It’s called the AI alignment problem, it’s fascinating, if you want to dig deeper in the subject I highly recommend ‘Robert miles AI safety’ channel on YouTube

    • @nevemsenki
      link
      31 month ago

      Computers do what people tell them to do, not what people want.

    • @Buddahriffic
      link
      31 month ago

      I read about a military AI that would put its objectives before anything else (like casualties) and do things like select nuclear strikes for all missions that involved destruction of targets. So they adjusted it to allow a human operator to veto strategies, in the simulation this was done via a communications tower. The AI apparently figured out that it could pick the strategy it wanted without veto if it just destroyed the communications tower before it made that selection.

      Though take it with a grain of salt because the military denied the story was accurate. Which could mean it wasn’t true or it could mean they didn’t want the public to believe it was true. Though it does sound a bit too human-like for it to pass my sniff test (an AI wouldn’t really care that its strategies get vetoed), but it’s an amusing anecdote.

    • @[email protected]
      link
      fedilink
      21 month ago

      ai thinks

      AI’s are Mathematic’s calculations. If you ordered that execution, are you responsible for the death? It happened because you didn’t write instructions well enough; test check against that which doesn’t throw life on the scale; or maybe that’s just the cheeky excuse to be used when people start dying before enough haven’t done so that no one is left A.S. may do it, if your lucky. Doesn’t matter. It’ll just bump over from any of its thousand T-ultiverses.