027: The AI who loved me

These days, everybody is talking about „The Spy who loved AI“. I have nothing to say about this topic. But I want to seize the opportunity to sort out some of my thoughts regarding „AI safety“ and „alignment“. As I understand, today the experts are roughly split into three camps. The first group is convinced that creating a superintelligence (ASI) is a very dangerous thing to do and that we must therefore be extremely careful when doing so. The second group agrees on the dangers, but believes that creating safe superintelligence is fundamentally impossible and that we therefore should not even try to create ASI. The third group on the other hand refuses to be scared and sees no reason why an ASI would want to kill us. They rather envision a bright future in which AI is bringing prosperity to all humankind. I think all three groups are - somehow - right. Let me explain why: Let’s start with the first group. It seems to be quite obvious that controlling a superintelligence won’t be…

Commenting on Bubbles has been removed. Why? If you want to talk about this post, write to the author, use the comments on their blog if it has them, or reply with a "RE:" post on your own blog. Suggestions? feedback@bubbles.town