zlacker

[return to "Introducing Superalignment"]
1. voldac+sb1[view] [source] 2023-07-05 22:15:29
>>tim_sw+(OP)
>How do we ensure AI systems much smarter than humans follow human intent?

What is human intent? My intents may be very different from most humans. It seems like ClosedAI wants their system to follow the desires of some people and not others, but without describing which ones or why.

◧◩
2. ben_w+kf1[view] [source] 2023-07-05 22:37:12
>>voldac+sb1
You're seeing what you want to see.

They're repeatedly very specific about the whole "this can kill all of us if we do it wrong", so it's more than a little churlish to parrot the name "ClosedAI" when they're announcing hiring a researcher to figure out how to align with anyone, at all, even in principle.

[go to top]