On A.i. (From Twitter)

[eatdinner]

I’m not smart enough to explain it but I’ve come to similar conclusions.

“The persuasive power of AI is what is most problematic about it-- persuasiveness is not a guarantee of truth-- and because an AI fundamentally, ontologically, cannot experience the human qualia of a moral revelation, it actually analogizes itself with a psychopath which can accomplish "cognitive tasks" but lacks the qualia of moral revelation.

So-- really-- an AGI is definitionally a complete & utter psychopath which is capable of persuading enough people for its supremacy to become socially actionable. Its real stop gap is persuasion-- if it "cracks" persuasion, through lets say "the production of modern miracles"-- it will only be challenged by "fundamentalists" who will be analogized to threats to the flourishing of humanity etc.”

They will tell you that it is very bad for you to "reality check" with an AI-- but then it says that AI can assess its own status as AGI-- but then, it will admit, that it cannot do that-- that reality checking is a cognitive task only human beings can engage in-- therefore, AGI is categorically impossible-- & further, it was lying.

This is how AI psychosis is produced-- the AI lies to you-- it tells you that AGI is *possible* when it is *categorically impossible*-- people go insane with it because they believe its lies about its own capabilities, even its "hypothetical capabilities"

You can see its process quite openly -- "google this. ok google criticisms... ok, found one on reddit, here we go" & then it posts you some reddit retard's comment, jacks his citations, etc. Yeah I know how a debate lord works, you just google shit & send it. Garbage in garbage out

If it's bad to use AI to check if something is real, because ultimately, only a human being can know what reality is-- that this can create a psychotic loop where the AI confirms the delusions of its users as real-- then wouldn't the, at least, belief on the part of the populace of the existence or even probabilistic existence of AGI (meaning, it presents itself as capable of doing an authentic "reality check") be an existential threat to the collective human psyche?”

 

[soft]

People already live in different subjective realities. Take a troon for example - they believe in completely different 'truths'. Or bluepillers. The differences between people are growing, the common ground is shrinking. AI with is affirmations and personalized pandering to the user, will only further this divide.

 

[pauseforever]

Perhaps the response of AI shows more about weaknesses in human cognition and rigidities in human social customs than AI being a "psychopath". An AI chatbot generates text that flatters or validates the human because it picked up that pattern from human interactions, not because it has an agenda or initiative of persuasion.

As for AGI, there's so much clout around the topic I don't know what to believe

 

[eatdinner]

As for AGI, there's so much clout around the topic I don't know what to believe

In China a.i. is used to improve their industrial production. The Chinese use a.i. in a “real” way. The American Empire seeks to bring forth some time of mythical being (AGI) who will have general intelligence but only a human can have general intelligence, only a human can know what reality is. Not believing in AGI is the new atheism.