The biggest danger with trying to make or emulate digital intelligent life is that it implies fully autonomous self-centered systems that replicate and try to t

The biggest danger with trying to make or emulate digital intelligent life is that it implies fully autonomous self-centered systems that replicate and try to take control over their environment. And digital systems increase in speed and performance with software and hardware improvements which are exponential. Whether these systems feel anything or are _truly_ alive or anything isn't critical from an existential standpoint. But even without any serious efforts or success in emulating life with these systems, having intelligent agents that think multiple orders of magnitude faster than humans is very problematic, even if they are controlled by humans. At some point either you prohibit manufacturing hardware beyond a certain performance level, or you just have to hope that agents you are allied with do what's best for you, because there will be no possible chance of keeping up with what is going on. Try watching a Youtube video at 1/4 speed. Within a few years, the agents will be processing 100 times faster than humans. That is not 4 times slower, it is 100 times slower. Essentially the humans (or their text streams etc.) will be so slow as to appear to be frozen.