I definitely don't think any computer developed "abstract thinking" on its own. Computers behave according to the programs that they are given by humans.
They did though. Learning software is how the chess challenge happened. But the point isn' a one to one comparison of development of computers and brains. It's exploring fundamental divisions between human thinking and computers, if they exist at all. So far, many of the things we've previously thought were special and unique about the human brain have been shown to be not so much.
I wouldn' be surprised if the 'ceilings becomes something more and more like the 'ceilings' creationists believe there is between micro and macro evolution. This thing which computers will never be able to do because there's something uniquely special about the human brain.
Is this post to me? How about answering these questions: How does the materialist justify the claims (1) that it is better to conform one's behavior according to certain moral precepts (e.g., that one ought to avoid causing unnecessary suffering), and (2) that it is possible to choose to conform one's behavior to any moral precept?
How do you? What non-material aspect is required? Once again, not all non-materislists believe
objective moral precepts exist, but in no way take that to me no moral precepts exist, they're chosen and influenced on an individual basis based on one's experience and upbringing. (Meaning both choice and nonchoice aspects are involved)
And some materialists do believe in objective moral standards, (usualy empiricistd along the scientism persuasion) and that people who don't look to maximize happiness while reduce suffering are biologically damaged.
I'm not one of them, thinking that empathy, while a useful evolved trait for coheive social structure, is no more natural than its absense, as there is no natural ideals. Nature doesn't care what we do. But we care what we do.