@AndyD273 In the same vein: Asimov came up with the three laws of robotics. In short, they are: 1. don't hurt humans 2. follow orders from humans 3. protect yourself from harm. Each following can be overridden by a previous one. So a robot won't jump in a volcano but if ordered by a human they would. But an order of "shoot him!" won't be followed. So, this is a rather decent and reasonable set of laws designed to prevent problems.

Well, most of Asimov's stories that use the three laws of robotics are about how those can go wrong in various ways due to conflicts between them. One story I recall was where robots were tasked with carrying some humans to a base or something. I think the humans were injured. However, the place had become irradiated, so the robots circled around in order to not harm the people they were carrying. Yet that went for so long, the humans died of exposure.

Asimov was spot on with the laws and robotics. He imagined machines capable of understanding and following complex orders. And yet still failing due to blind spots in all that reasoning. Which is the old "computers do what you tell them, not what you mean". Only it wasn't even old when Asimov started writing about robots.

@VLAZ Yeah, I think most of the Asimov 3 laws stories were about how easy it is to go bad with those things. Even the I Robot movie spiritually based on those stories followed that theme. That's why it's kind of funny when people suggest the 3 laws as a way to solve AI problems, when the whole point was that it doesn't.