Will AI Be The End of Everything?

Head to www.squarespace.com/drbenmiles to save 10% off your first purchase of a website or domain using code drbenmiles

Well, the biggest problem is, no matter how hard you try to regulate and restrict, there will always be that one overlooked person who will open the box.

He wrote those three laws, then stories about how it didn't work... few people seem to have read them.

8:22 "Building the System with American values" AI response? "I'll only be a dictator on the first day". Better be careful you vote for!

Iron rule of history: If it can be done, it will be done. This is strictly analogous to the "prisoner's dilemma." No avoiding the lower right hand corner.

this 3 laws need to be applied to all humans first ..

"...not less biased, just more moral"

What's another scary feeling is what if various AIs go to war with each other, and we are caught in the middle. Or even more ironically, they use humans as weapons.

You really think China, North Korea, Russia and Iran are going to leave the lid on Pandora's box? All of them? I don't.

The problem with AI originates in humans creating it. AI by itself is not good or bad, because it has no actual independent intellect, it is simply a mirror of its creators.

I love Isaac Asimov's writing. His Three Laws of Robotics weren't meant to be realistic, though. Their purpose was to stimulate interesting ideas for stories, as his characters continually tried to think of edge cases and ways around them - but the idea of instilling a very basic "ethical operating system" in every AI is a good one. If they were all programmed to do good at the most basic level of code, it might head off all sorts of situations where people try to misuse them. In that case their increasing levels of intelligence would work in our favor, since they'd be increasingly better able to understand "good" uses and refuse to do "bad" ones, even when we don't understand why it won't do what we say. But we'd have to find some way to make sure the AI didn't reinterpret good beyond the immediate situation (IE: is this command the user is giving me "good" for him and other humans right now), and don't develop some sort of moral philosophy where they only work for the good of the human species. No ends justify the means thinking for our flying weapons platforms, please!

It would be interesting to analyze if nuclear deterrence is a good way to preserve freedom and peace? 🤔

To prevent the kind of scenario your video alludes to, Asimov later added the “Zeroth Law,” above all the others – “A robot may not harm humanity, or, by inaction, allow humanity to come to harm.” Recall that in "I, Robot", the first three laws weren't enough, as the robot decided to come up with its own definition of "human"to circumvent them. So we'd better make sure the AIs are all in the same page in that regard

The three laws I believe have been scrapped as a concept. People may want to watch 'Robert Miles AI Safety ' on YouTube for a more in depth look at AI.

Military AI, at some form, is in operation since the 70s. Since things didn't end yet, I think, there's little reason to be worried.

Sending humans to war is cruel and a waste of lives. Sending robots to war is a waste of resources. Wars are nothing but waste trying to compensate for the lack of diplomacy or the lack of respect for people's rights. Wars are evil; and the poor pay the higher price.

Warfare via AI is just murder, not that war in general is that far off to begin with.

Stanislav Petrov is a.k.a. 'The man who saved the world'. 🙂

The only thing I dislike about the advent of AI is when people in your company who are responsible for AI systems are completely incompetent and ask for more and more human input from the other teams and wasting time for everybody. And their boss has no idea that it is unnecessary. Seriously, at my old job, there was a project ongoing for 3 years with a team. I could have finished it in 3 months without wasting time for everybody

The Swiss Author, Friedrich Dürrenmatt, wrote in 1962, "What was once thought can never be unthought". All the "limiting AI" talk is just that - talk. The World has turned dystopian long ago, and this is another example of "it needs to get worse before it will get better". If anything will be left to get better.