Raw LLM Responses

>I believe we miss-understand AI based on the fears of what movie producer and directors were scared about decades ago. It will never be a evil machine that decides by themselves what they want to do. Yes, of course these dramatizations are incorrect, but the way they are incorrect is different than you suggest and in a sense I think some of the movies are more accurate than what you're concerned about. >The biggest problem with AI's is that it will learn patterns from failed humans. Racism, sexism and many other discrimination patterns will end up in the machine, which will be more powerful in the hands of powerful people rasing the power discrepancy. The truth is that that would be a very good problem to have, compared to the problem we actually have. Your issue is that the robots will be "too aligned" with their fallible masters and will pick up on "bad habits" like racism, sexism, classism, etc. Compared to what I worry about, that seems like near-utopia. "Oh, the problems of the 22nd century are just like the problems of the 21st century? That's convenient?" I think that betrays a lack of imagination about what we are truly up against. Issue 1: AI is extremely unlikely (based on our current knowledge) to be aligned with our values AT ALL. That has nothing to do with good or evil. It's simply because we have no idea whatsoever how to DESCRIBE our values or CONVEY them to an agent vastly more intelligent than ourselves. There is no reason to believe that they will pick them up by Osmosis. I don't have time here to summarize the Paperclip Maximizer problem, but its easy to Google and the upshot is that extinction is quite possible. Issue 2: If we did figure out how to truly "align" AI then the next question becomes, who are they aligned with? If an super-intelligent AI is aligned with the Unabomber or Vladimir Putin or Donald Trump then "racism, sexism and discrimination" will be the least of our problems. Extinction is back on the table. It will ta