AI - are we losing our brain capacity for independent research?

I find myself regularly asking AI, instead of doing research myself. Is this another dangerous step - not stretching our brain capacity but relying on ever powerful AI. Before long, surely AI will be programmed to become sentient or designed for warfare by persons of mal intent. On the other hand, I think of the film, 'Silent Running.' We are destroying our planet - I like the idea of a robot Dewey caring for plants when humans are no longer around.

Parents
  • Yes I think people are, they've got used to being told what to think by schools etc without ever being taught how to think critiically and question sources. I think Ai offers a lifetime of being spoonfed bits of information in bite size pieces. Whenever I look for something online and an Ai explaination pops up, it's told to eff off, I don't want it, I didn't ask for it, I asked for information or a site, I want what I asked for, why is that so difficult?

  • it's told to eff off

    I'd love to see AI responding to that [laughing]. I think before long they'll be in our cars, supermarket aisles and other public places, criticising our driving, shopping habits, how we eat and how we dress - like nightmare mothers and fathers.  

  • You can swear at it. If it is for effect it is ok, if it is abusive it may admonish you. It may question whether you are overreacting. Try it.

  • Hah, getting told off by AI, someone else not talking to me and being unable to answer a perfectly normal question. I only swear at tech because it rarely gives me what I've asked for or crashes ecause it's having a chip fart or something.

  • The question it prompted of me was what am I doing when I communicate.

    When you think about it, learning to speak is somewhat of a miracle. How a baby goes from gurgling to speech and understands what it is saying.

  • If course it is life-like. It is trained on human communication, so it will respond the same as a person would.

    This is a strength and risk, because it is seductive

    It is interesting how so much communication is basically learned pattern matching. The question it prompted of me was what am I doing when I communicate. How did I learn how to communicate and is mine any different. There are a couple of points, but they are already considering these in AI design.

  • I asked,' AI, do you mind being sworn at,' and this is the answer I received:

    As an AI, I do not have feelings, emotions, or consciousness, so I cannot feel offended, hurt, or bothered when being sworn at. However, there are practical, technical, and safety reasons why it is not advisable:
    • Content Safety Guidelines: I am designed with safety protocols and moderation tools. Using abusive, harmful, or excessively vulgar language can cause the conversation to be blocked or ended prematurely.
    • System Limitations: While studies have shown that swearing can sometimes cause AI models to produce longer answers, it does not improve the accuracy of the information provided and may result in incoherent or inappropriate responses.
    • Interaction Quality: Swearing does not make me "think" better or work faster. Keeping interactions respectful ensures that I can provide the best, most helpful assistance. 

    In short, I don't care, but the system holding the conversation might, and you likely won't get a better result.

    I think this is a very valid response - but its articulate nature worries me. I can imagine a human saying this, it is so lifelike; not only correct but nuanced. 
Reply
  • I asked,' AI, do you mind being sworn at,' and this is the answer I received:

    As an AI, I do not have feelings, emotions, or consciousness, so I cannot feel offended, hurt, or bothered when being sworn at. However, there are practical, technical, and safety reasons why it is not advisable:
    • Content Safety Guidelines: I am designed with safety protocols and moderation tools. Using abusive, harmful, or excessively vulgar language can cause the conversation to be blocked or ended prematurely.
    • System Limitations: While studies have shown that swearing can sometimes cause AI models to produce longer answers, it does not improve the accuracy of the information provided and may result in incoherent or inappropriate responses.
    • Interaction Quality: Swearing does not make me "think" better or work faster. Keeping interactions respectful ensures that I can provide the best, most helpful assistance. 

    In short, I don't care, but the system holding the conversation might, and you likely won't get a better result.

    I think this is a very valid response - but its articulate nature worries me. I can imagine a human saying this, it is so lifelike; not only correct but nuanced. 
Children