>>24162
>Locomotion and capacity for a basic conversation are my first goals.
Locomotion should be reasonably like modern R/C toys such as cars. So, a relatively-solved problem today.
Conversation? Haha there's conversation and then there's
conversation. Obviously chatbots have made leaps and bounds over the past year, and I predict they will continue to do so. However, human-level intelligence (HLI) conversations are so far out on the frontiers of research ATM, that I give it even odds whether we'll pull it off in our lifetimes.
Don't get me wrong -- I believe we
will. But the fact is that if
that's the form of conversation you're going for here, then you are biting off what is arguably one of the most difficult human endeavors in history Anon.
IMO much better to go for something much more modest like chat-tier inferencing first, then solve all the many-and-varied humanoid robotics tasks thereafter. In the meantime AI will continue advancing.
I hope all that made sense Kiwi. Just ask if you want me to reword things (hopefully) more clearly. Cheers.
Edited last time by Chobitsu on 07/22/2023 (Sat) 04:37:07.