Nowadays we don't know how to create AI in a safe way (I think that we don't even know yet how to define a safe AI), but there is a lot of research in developing a model allowing it.

Let's say, that someday we discover such a model (maybe even it would be possible to mathematically prove its safety). Is it rational to ask, how do we prevent people from creating AI outside of this model (e.g. they are so confident in their own model, that they just pursue it and end up with something like paperclip scenario)?

Should we also think about creating some theory/infrastructure preventing such a scenario?

  • Did I understand you right, that you've recognized who wonderful a paperclip maximizer would be, because such a machine could produce unlimited amount of food, and build housing for everybody and you're asking here in the forum why nobody has invented such a great machine and how this matches to the promise of engineers to make the world a better place? – Manuel Rodriguez Aug 10 at 17:48
  • Of course I meant a bad scenario, when AI acts in an unforeseen way, that might be not possible to stop. – Ozaru Aug 10 at 19:42
  • "Nowadays we don't know how to create AI in a safe way" Nowadays we don't know how to create AI, dot. Current talks about rules to impose an unknown control model to unknown AIs are not empirical, not scientific, not true research. – pasaba por aqui Aug 11 at 11:48

It seems impossible to prevent that. If someone can make a safe AI from scratch in the near future, then someone else can make a dangerous AI from scratch as well. If all that's needed is a computer (or eventually a robot) it will be really hard to stop people from creating one.

Banning computers? Maybe it could prevent it, but that comes with quite a few negative sides as well.

A law against creating AIs? Would be really hard to follow-up, and what about the AIs people want to make to be used for something "good"?

And even if we could come up with a good set of laws they probably wouldn't be introduced in all countries, and also be really hard to follow up.

I guess we just need to make our own defence mechanisms which can fight unsafe AIs (physically or virtually) when that time comes.

Your Answer

 
discard

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.