I don't have any insight into the future. Anything might happen, including AGI. I doubt I will live to see it. I doubt my children or grandchildren will either.

I don't have any insight into the future. Anything might happen, including AGI. I doubt I will live to see it. I doubt my children or grandchildren will either. But I only have my own opinions informed by experience and what I read. I have worked in software development since the mid-1970s. I have see more fads and hype than I can list. I have seen advances in tech that have obvious and immediate value -- the internet, relational databases, personal computers, cloud computing. And I have seen plenty of other things that looked cool and got a lot of buzz but didn't work out (the dot-com bust, for example, or the numerous claims of "end user programming" aka no-code/low-code and now vibe-code). I try not to get too cynical but the hype around AI, with the blatant inflating of stock prices and crazy levels of infrastructure investment just looks too much like a scheme to fleece credulous and hopeful investors. One thing about working in tech that can turn out good or bad: we programmers frequently report to and get paid by people who not only can't do our jobs, they can't even begin to understand it. Our work looks like magic, or 10X intelligence. I get that from my parents every time I fix their router or get their PC to work again. Every once in a while the hucksters and parasites who see the tech industry as a gold mine get to cash in from investors and CEOs who don't understand the technology, and just gamble. Elon Musk's companies demonstrate how far someone can lie and make crazy promises for profit. I think the AI bubble will cost a lot more, and deliver a lot less. I tell myself I'll know AGI when I see it. I expect it might creep up on us rather than emerge fully-formed one day, if it happens at all. I like to keep an open mind about the possibility, but although I don't believe in souls or the supernatural I do think human intelligence and consciousness come from more than brain matter. I don't think human brains work like computers, even if we call some arrangement of tokens a "neural network." I think people can easily fall prey to scammers, and can delude themselves. Machines that can give a good imitation of humans might prove useful -- already have in some limited domains. I expect those domains to expand, just as automation has in the past. Some days I feel relieved that I likely won't live long enough to see AGI, and on other days I just want to stop hearing about it.