4 ms·
> than trying to push a concept of "alignment" that is not clearly defined or generally understood. True, I have no good idea of what people mean when they say
by JohnFen 4y ago
> than trying to push a concept of "alignment" that is not clearly defined or generally understood.
True, I have no good idea of what people mean when they say "alignment". But I also don't have a clear idea of what you mean when you say "symbiosis".
What is your definition of that in this context? What would it look like?
- zoroaster 4y agohttps://www.merriam-webster.com/dictionary/symbiosis https://www.merriam-webster.com/dictionary/symbiosis It is widely defined, established, and prevalent in nature. Pointing an AGI to "symbiosis" will give it a plethora of examples and ways to optimize for a symbiotic relationship with humanity and the planet. It is a much, much better optimizing function than some stupid human-centric "Alignment-problem" we are trying to solve for.
- JohnFen 4y agoI know what "symbiosis" means, just like I know what "alignment" means. I don't know exactly what you mean when you're using it in this context, though, like I don't know what others really mean when they use "alignment". What would this kind of symbiosis look like?