The way I see it, the likely, big ethical issue with AI isn't some Terminator/Butlerian Jihad scenario, or even mass surveillance - it's that the wealth created

The way I see it, the likely, big ethical issue with AI isn't some Terminator/Butlerian Jihad scenario, or even mass surveillance - it's that the wealth created by the technology will benefit the already wealthy much, much more than everyone else. This is generally true of most technology and even more so of the software industry (high scalability, "low" headcount, IP makes profit shifting and tax evasion trivial). But AI is unique in potentially enabling mass unemployment at a rate and scale never seen before in human history. This should be a joyous event, since humanity is finally free to enjoy the fruit of its labor without the labor part - but the way the world works today means most of us would not be allowed any bites of that fruit anymore. When Bezos fires every single warehouse employee, what happens if the job they start retraining for also gets automated away before they can even start? And the next one, and the next one. If nobody is making a salary anymore, then it doesn't matter how much lower the prices are on Amazon (due to being produced in automated factories and shipped from automated warehouses) unless Jeff decides to reduce those prices all the way down to 'free'. At that point the assumptions underlying the world's economy would break down in a way that makes a corona shutdown look like a mild hiccup. No software engineer is going to be able to do anything to help alleviate this. If you want to do something about this, you need to go into politics, not tech.