Raw LLM Responses

Your suggestion at the end that people need to watch more science and tech journals is light hearted and good spirited but I feel that the issue with awareness and education as science advances is not getting enough attention. The advancement in technology is in part due to extremely specialized sciences, expecting a layman to first have a motivation to understand it and second actually understand it is a tall order. For example, the summary they'll get after watching dozens of videos on LLMs is that it is 'probabilistic reasoning' and there are 'nodes' and 'weights'. Which is true, but I don't really see that as understanding, but as faith, until you verify it for yourself or see it demonstrated to be as such. A layman cannot really pop the hood of the models and see the inner workings, and with AI this is especially hard to do. So I don't find it surprising that stories are like this are coming out. To be fair, I wouldn't really blame it all on awarenees and education. Science itself doesn't really have a satisfying answer for if a thing is sentient or not. The fact that there are philosophical scenarios like the Chinese Room to demonstrate these dilemmas attest to this. Even in the Chinese Room scenario there is an argument to be made that the system is sentient even if the computer doing the dictionary lookups inside isn't. And people tend to naturally apply Occam's razor too. If it looks like a duck, quacks like a duck... So I would not put blame an individual person for getting ChatGPT induced psychosis, but rather hold the governing systems and creators who have released it freely accountable. Note: For people thinking that these people are dumb, there are some very rational people indulging in AGI cults in forums like LessWrong.