The feasibility of artificial consciousness through the lens of neuroscience
arxiv.org
arxiv.org
AI models are very simple compared to us, but it's interesting how the generalizations they come up with as a result of training appear to evolve just as smoothly, in seemingly useless parts that over time amplify each other and provide a generalized solution.
https://www.lesswrong.com/posts/N6WM6hs7RQMKDhYjB/a-mechanis...
You can say that self-awareness is the ultimate generalized solution for survival in a complex environment.