Posts
Comments
I'm not the best with this but I've been thinking of either an origin network seed or a spider like secondary network. The origin is something that can be tested off of, I guess similarly to a reporter. But unlike a reporter evolving we stunt the growth by saving this origin and then just passing through one sample then killing it. The way this could function goes into my second thought of spider, which has legs that sense things, a leg goes into a sensory object which leads to the main body. Similarly we can slowly stunt the growth so it learns slower and cannot become bad as fast as the central one(when it happens). Through the inputs there are many smaller hairs or flags, from how oddly this secondary network is feeling on a leg or how many flags are tripped in anomalous ways, it would require either another evaluation or origin rebooting. And if the origin is rebooted and still same thing occurs then either the entire AI has gone bad or the undesired outcome has occurred.