NO SYSTEM can evaluate its own output.
It certainly can, and it does, AI depends on this. Its output also becomes its own input.
Even a simple oscillator has this capbility. All radio systems and computer systems depend on it.
I have my computer evaluate its own logs.
TO Gurantee reliability, output must be evaluated by a different system ... and on and on infintely recursively.
Nope. Computers evaluate their own output quite regularly. This happens in every operating system.
And AI cannot even evaluate the outputs of its own sub actions.
It certainly can, and does.
IT gets dumber and dumber the more it goes because it takes is OWN AI SLOP AS TRUTH.
IT is not AI.
AI does not take it's own slop as 'the truth'. It is a program that simply re-evaluates its own slop with other information coming in.
Programs like Grok and ChatGPT echo the slop that's on the web, because they gather much of their information from there.
AI generates so much output it can not be realistically evaluated by humans.
Oddly enough, it is humans that programmed it, and supply ALL the data for its learning loop. Humans built the web, after all.
Yes there is.
Use AI as the useful too it is. It is not a Holy Oracle. It was never meant to be. It is not a Universal Truth. It was never meant to be.
Plus AI is taught to lie by training it on propaganda.
AI is not sentient.
whole movement fucked.
all supporters stupid.
AI is a useful tool, but it is not a Holy Oracle and not a Universal Truth. It is not sentient.
It is good at correlating navigation data. It is good at correlating what at first appears to be unrelated data.
It is getting better and better and correlating a crude understanding of English (and other languages). Indeed, it is better at English than most people here on JPP.
It is a lousy substitute for a brain. It cannot reason. Cut and pasting AI output is just spam and slop. It is not even presenting an argument. It's just mindless cut and paste for the extremely lazy.