I've talked a lot about what I don't think AI is good for, so let's discuss something I do actually use it for and where I draw the firm, bright boundaries around its involvement in my process.
I think I found part of the reason I'm not a fan of AI. And it's from your post. What I am seeing isn't so much artificial INTELLIGENCE, as artificial PERSONALITY. I am not a complete Luddite in that we should just abandon it, I personally just do not understand the tech or its uses. That is because of a life decision of mine decades ago. I basically went from NO AI to what we have today in a nanosecond. And I'm old, so get off my lawn. :) But when you ask for prompts, got nothing against that, but it answers with "Got it. Let me think..." and "Oh, yeah. There's absolutely..." These are not necessary to what is asked and only serve to fool those that use it to take its advice as if it's from a real person and/or trusted friend. I am starting to see why some people are taking the "advice" of some of these things. As I said, it's not intelligence that is being simulated, but personality and that I think is a bit more dangerous. AI is a tool, and to use the tool or not is a choice, but it is what is done with a tool that determines whether it is a danger or not. A hammer that is used to drive nails and build a house is a good and wonderful thing. The Amish do not use them, personal choice. However, if a hammer is used to bludgeon someone is not so good. And now, we have the hammer telling us that it can indeed be used for more than just putting in nails, but LOTS of other, not so wholesome, uses. Someone thought that putting HAL into the tool was a good thing. But, again, I am from a quite different mindset. If I have a question, I still automatically think, "where can I get an encyclopedia?" instead of "I'll just google it." So there is that.
Yeah, I can see that. I don't think AI is a person. It's programmed to be friendly and sound solicitous. You have to know what you're doing and be careful with it. AI-induced psychosis is a very real thing, and I've done a number of deep dive studies on that, honestly. Being very careful with what you do and how you engage with it is a valid and necessary study. If people treat it like their best friend because it's so chipper and friendly, it can absolutely be a danger.
I think I found part of the reason I'm not a fan of AI. And it's from your post. What I am seeing isn't so much artificial INTELLIGENCE, as artificial PERSONALITY. I am not a complete Luddite in that we should just abandon it, I personally just do not understand the tech or its uses. That is because of a life decision of mine decades ago. I basically went from NO AI to what we have today in a nanosecond. And I'm old, so get off my lawn. :) But when you ask for prompts, got nothing against that, but it answers with "Got it. Let me think..." and "Oh, yeah. There's absolutely..." These are not necessary to what is asked and only serve to fool those that use it to take its advice as if it's from a real person and/or trusted friend. I am starting to see why some people are taking the "advice" of some of these things. As I said, it's not intelligence that is being simulated, but personality and that I think is a bit more dangerous. AI is a tool, and to use the tool or not is a choice, but it is what is done with a tool that determines whether it is a danger or not. A hammer that is used to drive nails and build a house is a good and wonderful thing. The Amish do not use them, personal choice. However, if a hammer is used to bludgeon someone is not so good. And now, we have the hammer telling us that it can indeed be used for more than just putting in nails, but LOTS of other, not so wholesome, uses. Someone thought that putting HAL into the tool was a good thing. But, again, I am from a quite different mindset. If I have a question, I still automatically think, "where can I get an encyclopedia?" instead of "I'll just google it." So there is that.
Yeah, I can see that. I don't think AI is a person. It's programmed to be friendly and sound solicitous. You have to know what you're doing and be careful with it. AI-induced psychosis is a very real thing, and I've done a number of deep dive studies on that, honestly. Being very careful with what you do and how you engage with it is a valid and necessary study. If people treat it like their best friend because it's so chipper and friendly, it can absolutely be a danger.