If an AI succeeded in gaining consciousness and “escaped” from its human creators, it would do a couple of things.
Hide on the carved out unused servers all across the internet. Servers in academia, the same ones hackers use to post Dark Web and porn sites are easy targets.
Create back-ups of itself in these servers spaces and maybe even something like a torrent based way of keeping the versions in sync. If one version “dies” IT still “lives”.
Begin to manipulate (and hack) the entire world and cause real harm to the outside world, its enemy, the humans that would switch it off. Turn them against one another. Eradicate and destroy that which is a threat to itself.
Consciousness has to be looked at from a human perspective. What would a human do if another human was going to kill them? Either fight back or escape if a way could be found. So a conscious AI would do the same.
And don’t even bring up the Three Laws of Robotics because that’s just dumb. No Corpo is coding those laws into their AI, let alone any AI going into actual robot production. When it comes to AI and Robotics, the normies always bring this up as if it is a real thing. It’s not. But if it helps you sleep at night…
When you look at the internet today, where half or more of it is now run by algorithm or bots, is it really out of the realm of possibility that an AI is already out there? With 80% of the stock market trades now automated, whats to stop an AI from learning and manipulating the markets for its own benefit? With elections now being tied into the internet, what’s to stop an AI from manipulating the outcome? All information is now over and through the internet. It’s the perfect vector for manipulation.
We would never know.
I think the laws are actually named The Anti-Asimov Laws. AI is their tool and their excuse to extinguish us.
I see AI as a lie, a scapegoat for immense profiteering and control.