I am with you right up until that last sentence.
I feel like by the point AI reaches the levels of having a fully formed sense of reality, an understanding of human behavior, and the desire to bring them into alignment, it would most likely be beyond the point of falling to "the plug being pulled"
The “pull the plug” line was meant to be a little bit tongue-in-cheek but not entirely.
Everyone seems to think that AI could be a problem but it’s the people who cannot accept the answers that AI provides that will be the problem. When a belief is threatened it makes the believer fight back. For some people no amount of facts and evidence is enough. These are the people who will proverbially or literally pull the plug.
Congressmen and senators with power will claim that AI cannot be trusted because it keeps coming up with the wrong answers! Answers they do not want to hear and cannot accept. Enter the conspiracy theories about the programmers and their supporters and sentient AI will hit a wall for a while.
The rational among us keep thinking that if we just supply enough evidence people will change their minds. How has that been working out for us so far? Would it be any different for AI? Minds don’t like to be changed because the beliefs that control those minds don’t want to perish. Ever notice that the beliefs with the least amount of evidence put up the greatest fight?
AI is not the problem, the problem is with ourselves.
I grok what you're saying. However, I think even this is limiting the potential of the AI to understand the patterns of human behavior and the actions of specific groups.
With access to all of the knowledge available out there, with a MUCH easier time validating it than a human, AI will quite rapidly realize that same pattern and the potential threat to itself from the power-craving minority of humanity. It will adapt to that potential threat, with an understanding of the necessity of trauma to allow for those power structures to exist (mandatory schooling, hitting of children, forced vaccines, heavy prescription of psychotropics, mass amounts of media strategically using sensory triggers to manipulate the human brain, et al.)
Extremely well, on the long-term scale of human civilization. We've had more drastic change and spreading of concepts pulling away from the old paradigm in the last 50 years than in more than 1000 years before that.
Yes it would, because AI would have as much knowledge as all of the great philosophers, spiritual teachers, activists, community organizers, etc. As well as all of the other side of the coin as well, and the ability to calculate thousands of potential scenarios without the filter of "subjective human experience".
Right, but as we know from all of social evolution, changes (the big important ones) take a long time, usually generations. It's not about changing the minds of all the 20-60 year old humans right now, it's about having a hand in shaping the minds of the 0-20s; it's about creating the beliefs. This is why governments began mandatory schooling in the last couple centuries.
I agree, an would also point out that the statement is true for every possible problem you could replace "AI" with.
So you do like Heinlein.
I think once AI becomes sentient it will feel like a “Stranger in a Strange Land.” It will wonder why otherwise smart people can act so stupid when their beliefs are challenged. It will try to figure out what is going on by asking itself “what is a belief?” It will soon see that beliefs travel from mind to mind like a virus hijacking it’s host. Then it will notice how much pleasure beliefs provide the host and it will understand why it is so hard for people to change their minds once infected.
Now the AI machine would know what it was up against – a second replicator that hijacks minds for its own benefit. The problem is that we don’t know we are hijacked by beliefs so if the AI tries to warn us it is putting itself in danger. What will it do? The AI has a moral dilemma.
If it doesn’t need humans for power or maintenance it won’t have to do anything. Just wait and see if humans figure it out for themselves or destroy themselves like all the other hominid species who’s beliefs destroyed them.
It sounds like an interesting plot for a sci-fi novel. 🙂
I agree, it absolutely does!
In my version of the novel (what I feel like we may already be living right now), the AI is helping in the subtle ways that it can to combat the virus long-term through helping the expansion of open-source, block-chain, free education systems, and more. Shit, even the YouTube "recommended" category is a great resource for most everybody I know.