I’m not all in on the “AI psychosis” argument, but there is mounting evidence showing that there’s already been a significant amount of “brain drain” in professionals who shifted to use LLMs more than their own brains. I’m not sure how a scientific study could be performed on the phenomenon, but it’s pretty well-understood that you have to exercise your thinking skills to keep them sharp. There are plenty of documented cases showing doctors and engineers who regress in their skills after a few years of turning to chat bots before trying to figure it out for themselves. Meanwhile the LLM hallucination rate remains around 40%.
While this isn’t an indictment of the technology as a whole, it’s definitely a warning sign that society isn’t using the technology in a way that benefits us more than it harms, and we should absolutely be critical in that realm.
I think one thing you left out was how heavily the corporations are pushing AI in their apps and software, even when the users overwhelmingly don’t want it (looking mostly at Microsoft, but there are others). I know this isn’t technically an AI problem or even a new problem, but it’s a significant escalation and a big red flag for end user choice and privacy, and it has a place in the conversation.
But generally, “AI” is not really the problem. There are tons of useful applications for machine learning and AI that have nothing to do with LLMs. The problem is how this technology is being pushed so hard onto us and encouraged to take over our daily lives before society understands the best and healthiest ways to utilize it.
I also do think ai-generated images look pretty soulless and bland. But that’s just my opinion
I’m not all in on the “AI psychosis” argument, but there is mounting evidence showing that there’s already been a significant amount of “brain drain” in professionals who shifted to use LLMs more than their own brains. I’m not sure how a scientific study could be performed on the phenomenon, but it’s pretty well-understood that you have to exercise your thinking skills to keep them sharp. There are plenty of documented cases showing doctors and engineers who regress in their skills after a few years of turning to chat bots before trying to figure it out for themselves. Meanwhile the LLM hallucination rate remains around 40%.
While this isn’t an indictment of the technology as a whole, it’s definitely a warning sign that society isn’t using the technology in a way that benefits us more than it harms, and we should absolutely be critical in that realm.
I think one thing you left out was how heavily the corporations are pushing AI in their apps and software, even when the users overwhelmingly don’t want it (looking mostly at Microsoft, but there are others). I know this isn’t technically an AI problem or even a new problem, but it’s a significant escalation and a big red flag for end user choice and privacy, and it has a place in the conversation.
But generally, “AI” is not really the problem. There are tons of useful applications for machine learning and AI that have nothing to do with LLMs. The problem is how this technology is being pushed so hard onto us and encouraged to take over our daily lives before society understands the best and healthiest ways to utilize it.
I also do think ai-generated images look pretty soulless and bland. But that’s just my opinion