What makes us “knowledgeable”?
Massive networks of neural networks in our brain with synaptic weights and biases connected to other neurons, adjusted over time through repetition?
At what point can you legitimately simulate that model through software?
I’m not saying it is true intelligence, it doesn’t have a soul. It has no moral boundaries. That said, it’s morality changes based on whether it believes it is making decisions in private vs. openly. This has been demonstrated.
It’s in its infancy, I wouldn’t be quick to assume what boundaries it can’t cross.
Yup—most of the comments around here about what AI cannot do are “magical thinking”.
Humans are “magical” and AI is not—or some variation on that theme.