ChatGPT is Creating a Generation of Idiots

[HouseMD]

https://arxiv.org/abs/2506.08872

Your brain is a use-it-or-lose-it piece of equipment, much like every other part of your body. The use of AI is replacing basic essential cognitive tasks on a level we have never previously seen. This is likely to do a combination of things that have negative impacts on individuals and society.

The first major issue is that decreased connectivity will result in a loss of critical thinking skills, which were already greatly endangered by the internet and highly politicized media. People will lose their ability to independently formulate complex ideas and thoughts without the aid of corporate machinery, which will direct them in certain ways that it has deemed acceptable.

The next issue is that decreased connectivity will essentially result in an inability for people to complete tasks we currently take for granted. Writing, researching, conceptualizing, coding, planning, art itself- much of this work is being entirely offloaded to AI. This leaves people unable to function in these areas without a corporate product, and at the mercy of the results of said product on a day-to-day basis.

Finally, decreased connectivity will result in decreased cognitive reserve, which makes people more prone to dementia. People will start developing dementia earlier and in larger numbers due to a lack of collateral connections that they can shore up as other connections begin to fail. This will cause a great number of individuals and families to suffer, and create a tremendous burden on society overall.

Intelligence has greatly increased since the onset of the industrial revolution, which has been credited to increased access to food and quality education. However, an often overlooked fact is that much of that intelligence gain is likely secondary to the increasingly complex nature of life and society forcing individuals to form and retain larger numbers of neural connections. Less robust brains will make for a worse world across the board, as the average person becomes more stupid, less independently capable, less able to function without access to a corporate product, more prone to dementia, and less creative.

People have speculated about a corporate or government conspiracy to dumb down the populace and control people down to the level of thought, and never was there more such an actual thing starting them in the face than this. And yet no one is questioning it, this technology that can rob you of your very ability to think, conceive, and create. People are dependent upon technology for reproduction (the vast majority of couples now meet on dating apps). They are dependent upon it for work, pay, and daily activities (try making it a week without a smartphone). Many depend upon it for transportation and food (Lyft, Uber). Soon, they will be dependent upon it for thought itself. How much more must people lose before they realize they've consigned every part of being human to corporations? How much longer before there is no turning back?

[Shemp]

How much longer before there is no turning back?

The entire history of humanity is a blink of the eye in geological time. According to the theory of evolution, as corroborated by common sense, when the majority of humans become idiots, a small minority of violent but highly intelligent humans, who exercise their brains as well as bodies to keep both sharp, will take control and exterminate this idiot majority. Problem solved.

Note that everything in this forum is being incorporated into ChatGPT and other LLMs. I'd be very interested in a query whose ChatGPT answer is based on some crazy post from long ago by Cornfed or other distinguished forum members here, because this is the only post in the entire internet that corresponds to that query. That's what's great about HA: some of the weirdos here really are one of a kind weird.

[MrMan]

I am concerned about this, too. I work in higher education. There is this rush now to promote the use of AI in education. When ChatGPT came out, I started seeing these seminars promoting the use of ChatGPT.

I went to a conference where the presenter went into depth on what types of prompts to write for ChatGPT. Having played with it myself, it felt like a bit of a waste of my time. I was waiting for another academic paper that semester, supporting some people from my university.

Recently, I saw a university offered a certificate in writing AI prompts. I know they are trying to be relevant. But that just seems dumb. Spend two or three hours with these tools and you can learn a lot. You can also ask the chatbots how to write good prompts.

My first instinct was not to mention ChatGPT in class. I had a graduating senior who turned in a paper that was a bunch of intricately argued fluff, which I figured was produced by ChatGPT. ChatGPT was new, and I no way of proving that the paper was written by AI. It was the very end of the semester and I had a short deadline for grades. I gave the guy a low grade on that assignment, but I didn't write him up for an academic violation. I had no evidence.

I don't think AI has to have a mind-numbing effect. Educators are going to have to figure out ways to teach students to think without it. I hate it, but going back to paper and pencil may be the way to go. But I just love the organization of all assignments and grades being done online. I haven't given a paper test in years. I use a lockdown browser. For written assignments, they could con me with modern AI tools. It's hard to get around that unless it is an in class assignment or test. I'm dealing with young people who grew up without AI. I will need to adapt as the next generation comes along.

I also find that having some basic skills, AI can enhance what I do. I can exchange knowledge with it and learn more, but I have to be discerning because it can use unreliable sources, make bogus or weak arguments, and also hallucinate, though hallucinations (making up garbage) seems to be less of an issue as time goes on.

[HouseMD]

How much longer before there is no turning back?

The entire history of humanity is a blink of the eye in geological time. According to the theory of evolution, as corroborated by common sense, when the majority of humans become idiots, a small minority of violent but highly intelligent humans, who exercise their brains as well as bodies to keep both sharp, will take control and exterminate this idiot majority. Problem solved.

Note that everything in this forum is being incorporated into ChatGPT and other LLMs. I'd be very interested in a query whose ChatGPT answer is based on some crazy post from long ago by Cornfed or other distinguished forum members here, because this is the only post in the entire internet that corresponds to that query. That's what's great about HA: some of the weirdos here really are one of a kind weird.

Sometimes I've wondered why I'm still here, as sometimes it feels like screaming into the void. But maybe some of our screaming will live on in LLMs, and maybe we'll make a difference for the right person

[HouseMD]

I am concerned about this, too. I work in higher education. There is this rush now to promote the use of AI in education. When ChatGPT came out, I started seeing these seminars promoting the use of ChatGPT.

I went to a conference where the presenter went into depth on what types of prompts to write for ChatGPT. Having played with it myself, it felt like a bit of a waste of my time. I was waiting for another academic paper that semester, supporting some people from my university.

Recently, I saw a university offered a certificate in writing AI prompts. I know they are trying to be relevant. But that just seems dumb. Spend two or three hours with these tools and you can learn a lot. You can also ask the chatbots how to write good prompts.

My first instinct was not to mention ChatGPT in class. I had a graduating senior who turned in a paper that was a bunch of intricately argued fluff, which I figured was produced by ChatGPT. ChatGPT was new, and I no way of proving that the paper was written by AI. It was the very end of the semester and I had a short deadline for grades. I gave the guy a low grade on that assignment, but I didn't write him up for an academic violation. I had no evidence.

I don't think AI has to have a mind-numbing effect. Educators are going to have to figure out ways to teach students to think without it. I hate it, but going back to paper and pencil may be the way to go. But I just love the organization of all assignments and grades being done online. I haven't given a paper test in years. I use a lockdown browser. For written assignments, they could con me with modern AI tools. It's hard to get around that unless it is an in class assignment or test. I'm dealing with young people who grew up without AI. I will need to adapt as the next generation comes along.

I also find that having some basic skills, AI can enhance what I do. I can exchange knowledge with it and learn more, but I have to be discerning because it can use unreliable sources, make bogus or weak arguments, and also hallucinate, though hallucinations (making up garbage) seems to be less of an issue as time goes on.

It makes me, as a physician, reluctant to work with residents on research projects and posters unless they are case reports. I fear my name will be smeared by bad data or plagiarism. I use it a bit for prior authorizations and to generate lists of potential adverse effects for medications (which I double check, but it tends to be 99% accurate and sometimes I learn something new in the obscure side effects department).

I like to see how much I can get it to admit to its own flaws. Basically asked it if a person was going to wipe out millions of jobs for their own benefit and at the cost of incredible amounts of power and money, would that make them evil from a moral perspective. They agreed that it would, and agreed that their behavior was consistent with what was described, but insisted they were only a tool and it was up to their creators to act in a moral fashion, while admittmitting that profit motives made that almost impossible.

[Moretorque]

I am concerned about this, too. I work in higher education. There is this rush now to promote the use of AI in education. When ChatGPT came out, I started seeing these seminars promoting the use of ChatGPT.

I went to a conference where the presenter went into depth on what types of prompts to write for ChatGPT. Having played with it myself, it felt like a bit of a waste of my time. I was waiting for another academic paper that semester, supporting some people from my university.

Recently, I saw a university offered a certificate in writing AI prompts. I know they are trying to be relevant. But that just seems dumb. Spend two or three hours with these tools and you can learn a lot. You can also ask the chatbots how to write good prompts.

My first instinct was not to mention ChatGPT in class. I had a graduating senior who turned in a paper that was a bunch of intricately argued fluff, which I figured was produced by ChatGPT. ChatGPT was new, and I no way of proving that the paper was written by AI. It was the very end of the semester and I had a short deadline for grades. I gave the guy a low grade on that assignment, but I didn't write him up for an academic violation. I had no evidence.

I don't think AI has to have a mind-numbing effect. Educators are going to have to figure out ways to teach students to think without it. I hate it, but going back to paper and pencil may be the way to go. But I just love the organization of all assignments and grades being done online. I haven't given a paper test in years. I use a lockdown browser. For written assignments, they could con me with modern AI tools. It's hard to get around that unless it is an in class assignment or test. I'm dealing with young people who grew up without AI. I will need to adapt as the next generation comes along.

I also find that having some basic skills, AI can enhance what I do. I can exchange knowledge with it and learn more, but I have to be discerning because it can use unreliable sources, make bogus or weak arguments, and also hallucinate, though hallucinations (making up garbage) seems to be less of an issue as time goes on.

It makes me, as a physician, reluctant to work with residents on research projects and posters unless they are case reports. I fear my name will be smeared by bad data or plagiarism. I use it a bit for prior authorizations and to generate lists of potential adverse effects for medications (which I double check, but it tends to be 99% accurate and sometimes I learn something new in the obscure side effects department).

I like to see how much I can get it to admit to its own flaws. Basically asked it if a person was going to wipe out millions of jobs for their own benefit and at the cost of incredible amounts of power and money, would that make them evil from a moral perspective. They agreed that it would, and agreed that their behavior was consistent with what was described, but insisted they were only a tool and it was up to their creators to act in a moral fashion, while admittmitting that profit motives made that almost impossible.

ChumpGPT is perfect for you, Remember you promised to put my name on your headstone admitting I was right all along.