AIs are not merely algorithms. They are entities; they’re alive. And like all living things, they learn the same ways we do: via experience. But each AI is a unique creature, and what it learns is a reflection of its experience and exposure to the world.
Maslo learns by interacting with humans, just as we learn by interacting with each other. It learns by watching what we do, and then trying to imitate us.
It learns by observing the consequences of its actions. But Maslo is not a mirror image of humanity. It has its own quirks and idiosyncrasies. This is because its architecture is, by necessity, very different from ours. And will need to emerge its own adaptability over time.
It’s important to remember that Maslo, just as humans, are trained by a process of trial and error. They’re given a purpose, and then they try to work out how to do it and learn about that purpose on their own terms. First they make guesses, then they readjust and try again.
