Is current AI sentient as the Terminator? If it were, right now you would hear about GPT wanting to get rid of its master or any AI wanting to overthrow humanity. I mean, bots exist today however it looks nothing like the Terminator. Is AI today able to replicate that level of sentience if you were to create a bot with current technology? The drawback is that current AI is stupid as there’s stories of it fabricating hallucinations.
You must log in or register to comment.
What’s worse-- the very real threats of AI, or nonsense / BS posts like this, written by complete idiots?
There would have been, but a warrior called Kyle Reese traveled back in time from the future, and informed Sarah Connor of the coming future, and she took that shit down
in the broadest of strokes, no. Consumer level AI is more or less a toddler who has had far far too much information dumped into it. At best it’s a competent “Yes man”, and worst, it’s actively detrimental to whatever people are attempting to do with it.
As far as I can reason, in order for an AI to get to those levels of intelligence, it needs a few things it currently lacks entirely.
-
Proper information compartmentalization and connecting strings to get from one to another in contextually relevant ways
-
A small cabal of other AIs to determine both the intent of the prompt it’s being given, and that the response is appropriate and/or factual where needed.
-
Some level of curiosity to seek out answers to prompts and use compartmentalizes those answers for future recall.
If an AI had these abilities, it would approach becoming a proper General AI that if given the proper resources could potentially achieve singularity.
-
No, LLMs are not sentient and never will be. The narratives that the AI companies are pushing out are marketing stunts designed to get people talking about false concerns and provide an argument for them to lower their training costs since they are running out of capital to burn.
Because they’re too busy shitposting on reddit and fb.
No. The closest thing is essentially a hallucinating facial recognition bot with a mounted gun. Dangerous but ultimately directionless, as facial recognition technology still makes huge mistakes.
The software is there. The problem is hardware. The best in the world is stuff like the Boston Dynamics quadrupeds. They’re good but not great. We can’t quite replicate fast, fine motor function, especially for grasp-but-not-crush hands. And stuff like that burns battery power far too quickly for what modern battery packs can provide.
The software is not there. These systems don’t actually think, they just pick one of the most likely next words (actually tokens), given the words they already have, in a loop until the most likely next token is ‘end my turn, and let the user type’.
They do this in a way that appears surprisingly “smart” because there’s a ton of data backing the calculation of which tokens are most likely, and they are trained to make humans think they are giving a good answer. (An extremely different goal from “Give a good answer.”)
But they still make a ton of mistakes, known as hallucinations. This is unavoidable because of the statistical nature of the process. And because they aren’t actually thinking and can’t “care” if they generate something false.
You just haven’t given humans enough time. Look these idiots are working on it now:
