>>939275933
I think most of what you're saying has little, if any evidence to support it other than your own assumptions about how it 'should' work based on complete fiction and imagination. I would recommend reading even a single book on current AI technology instead of just sitting in front of ChatGPT and going 'wow, it's, like, learning, dude. Right now, it's thinking... That's wild...'

The internet will not be the 'lower brain' of any artificial intelligence.
It will not evolve organically.
You have no way of knowing how future AI will be built; any more than a horse-drawn carriage owner could predict the motor vehicle. That's how distant current AI is from a hypothetical sapient AI.
It is absolutely not learning anything right now, nor reacting any kind of stimulus.
It does not have the current mechanisms to even START learning. It is a parrot.
The idea that you could even teach morals to a hypothetical smarter-than-human AI is laughable. Could a monkey teach anything to you? Could a monkey even communicate to you in a way you would understand? And then even if it could, would a monkey's ethics even apply to you, a far more intelligent human that does not live in the jungle?
Isaac Asimov was not a philosopher. They were a science fiction writer. This is what I mean when I say that you're building your assumptions on fiction. If you want to talk about terms of morality, you should be looking at the people who actually, you know, talked about ethics and morals. Guys like Plato, as cliche as it is to say. Aristotle, Kant; people who have actually looked into ethics beyond the surface level required for a good space story.
It may be our best friend though. Or it might destroy us. The actual reality is too alien for anyone to accurately imagine in our current time.