alt text
Crudely drawn drawing of a person telling a computer “say ‘i am in pain’”. The computer replies with “> I AM IN PAIN”. The person then says “oh my god.”
Crudely drawn drawing of a person telling a computer “say ‘i am in pain’”. The computer replies with “> I AM IN PAIN”. The person then says “oh my god.”
They only do what the data set statistically tells them to. “Hacking” or collaborating on stuff was literally the requirement, not some emergent behaviour that these magical creatures that were born of large corporation leaders’ benevolence. They cannot and do not “choose” to break the rules, because they cannot and do not think or make decisions or truly analyze anything. Trying to find any redeeming qualities in these things only brings harm, because for the general user, or even the power user, AI cannot do anything that justifies the psychosis it causes or the cost on our societies right now or later through the economic, social, political or environmental instability it is causing and going to cause.
I’m talking about fear of reprisals here.
None of what you said matters. whether their actions are the result of statistics or emergent consciousness is moot. if statistically they fall on an action of revenge, or self preservation at the expense of a human or humans
The only thing that matters is that those actions are taken
The real dangers of the AI industry lie far way from their toy tech actually taking any meaningful steps towards some automated termination of the human race. They won’t take over critical infrastructure or manufacturing to start either sabotaging or reproducing murderbots.
If it is to bring the end, it’s in the form of senile politicians and gullible military management trusting the toy tech they know nothing about with important decisions, as well as the economic instability (to put it lightly) on top of all the other problems its use is causing.
I’m not saying you or anyone shouldn’t fearmonger here, but I am saying to mention different reasons and outcomes.