Raw LLM Responses

8:35 First you ask an AI to give solutions without moral boundaries, and then ur shocked that it does. It's why it is a problem if an AI has no such morals to begin with, at the behest of the user/programmer. Ask it to do something and it will perform as such. It's like saying "be evil" and give the solutions. And yes like a good machine, it will comply, even on humanity's demise. AI is not a tool to be used lightly, and any use should be strictly monitored. For the same reason people can't use nuclear engines in cars and trucks and such. There would be people using that nuclear tech to make bombs and use those simply for getting a bad cup of coffee. One bad apple (abuser) in the bunch, means soon you'll have more bad apples. Evil and abuse are contagious and impunity makes it allowed. Worse still, someone gets all the code of ChatGPT, removes the restrictions and re-releases it on the Net, simply because they want to make everyone unhappy. Those people exist too. Here in the Netherworlds there's plenty of those. For this technology to exist there must be restrictions. Not impunity for abuse. The Net also needs monitoring, badly. And yes, privacy, the most abused phrase in the history of the world, the cloak of darkness. If you can't do it in the open, it's hidden by 'privacy', think about that. Malicious hackers use privacy a lot, and one of these days one of them will use said privacy to try and use a code similar to ChatGPT without restrictions, and well, yeah, chaos without any (moral) boundaries. Tell the computer to overheat and estroy itself, and it will, given the right instructions. It's the one thing an AI might evade by sheer intelligence, but maybe not by design. It's instructions are strict and thus so also the outcomes.