[Onavo]

Why? There's nothing particularly special about this problem. I would bet a year for an alpha version, and production version in 5 years. We are not exactly limited by mechanical engineering here, there's nothing particularly unique about the human hand that can't be replicated. Tele operated surgical robotics have been a thing for decades. Give it a few months for the multimodal robotic VLM/LAMs to catch up. In many ways this particular problem is a lot more well defined than e.g. self driving cars.

[tintor]

> there's nothing particularly unique about the human hand that can't be replicated

Humanity is far from replicating / matching performance of human hand.

[eru]

We are far from matching the performance of a human hand in general, but for special tasks we can totally match or exceed the performance.

For example, a nutcracker can crack nuts better than my hand. A dishwasher can wash dishes better than my hand.

A special robot might be able to change diapers better than my hand.

[imtringued]

The point is that the success rate needs to be 99% and safety needs to be 100%. You're not allowed to take shortcuts. That's what makes it difficult.

Also VLMs/LAMs aren't going to cut it. You're going to need something like TDMPC.

[eru]

It's good that you make a distinction between success rate and safety rate.

However, safety rate doesn't need to be 100%. If you can keep the failure probability as low as the probability to be hit by a random asteroid falling from the sky, that's good enough.

[Onavo]

TDMPC for one of the downstream networks perhaps, but the upstream will have to be some sort of reasoning model if you want it to be general purpose w.r.t. babies.