Hacker News Reader

Comment by Animats | Hacker News Reader

There are at least 18 humanoid robots good enough to have Youtube videos of them moving around. Some are far more agile than this one.

Needs more manipulation. Such elaborate fingers and all it does is mime carrying a box. There are some brief material handling demos at the end, but nothing challenging.

There's been considerable progress in robot manipulation in the past year, after many decades of very slow progress. This year's new manipulation demos have been for fixed base robot hands. Robot manipulation still isn't good enough for Amazon's bin picking. The best demo of 2025 is two robot hands opening a padlock with a key, with one hand holding the lock while the other uses the key.

We'll probably see this start to come together in 2026.

View on HN

> We'll probably see this start to come together in 2026.

Thus far I see no evidence that robot manipulation will come together by 2036, let alone 2026.

I dunno. The folks at Physical Intelligence are showing remarkable progress for being such a small team and relying on Gemma as their base model.

https://www.pi.website/blog/pistar06 has some reasonable footage of making espresso drinks, folding cardboard boxes, etc.

That's what I was thinking, but could not find the link. Here is it working on some standard tasks.[1] Grasping the padlock and inserting the key is impressive. I've seen key-in-lock before, done painfully slowly. Finally, it's working.

That system, coupled to one of the humanoids for mobility, could be quite useful. A near term use case might be in CNC machining centers. CNC machine tools now work well enough on their own that some shops run them all night. They use replaceable cutting tools which are held in standard tool holders. Someone has to regularly replace the cutting tools with fresh ones, which limits how long you can run unattended. So a robot able to change tool holders during the night would be useful in production plants.

See [2], which is a US-based company that makes molds for injection molding, something the US supposedly doesn't do any more. They have people on day shift, but the machines run all night and on weekends. To do that, they have to have refrigerator-sized units with tools on turntables, and conveyors and stackers for workplace pallets. A humanoid robot might be simpler than all the support machinery required to feed the CNC machines for unattended operation.

[1] https://www.pi.website/blog/olympics

[2] https://www.youtube.com/watch?v=suVhnA1c7vE

Most flexible manufacturing systems come with a central tool storage (1000+ tools) that can load each individual machine's magazine (usually less than 64 tools per machine). The solution to the problem you mention is adding one more non-humanoid machine. The only difference is that this new machine won't consume the tools and instead just swaps the inserts.

There is literally no point in having a humanoid here. The primary reason you'd want a human here is that hiring a human to swap tools is extremely cost effective since they don't actually need to have any knowledge of operating the machines and just need to be trained on that one particular task.

> A humanoid robot might be simpler than all the support machinery required to feed the CNC machines for unattended operation.

A humanoid robot is significantly more complicated than any CNC. Even with multi-axis, tool change, and pallet feeding these CNC robots are simpler in both control and environment.

These robots don't produce a piece by thinking about how their tools will affect the piece, they produce it by cycling though fixed commands with all of the intelligence of the design determined by the manufacturer before the operations.

These are also highly controlled environments. The kind of things they have to detect and respond to are tool breakage, over torque, etc. And they respond to those mainly by picking a new tool.

The gulf between humanoid robotics in uncontrolled environments is vast even compared to advanced CNC machines like these (which are awesome). Uncontrolled robotics is a completely different domain, akin to solving computation in P by a rote algorithm, vs excellent approximations in NP by trained ML/heuristic methods. Like saying any sorting algorithm may be more complex than a SOTA LLM.

I think manipulation will come long before 2036, but the people doing high level planning on LLMs trained on forum discussions of Chucky movies and all kinds of worse stuff and planning for home robot deployment soon I think are off by a lot. Things like random stuff playing on TV rehydrating that memory that was mostly wiped out in RLHF; it will need many extra safety layers.

And even if it isn't just doing crazy intentional-seeming horror stuff, we're still a good ways off from passing the safely make a cup of coffee in a random house without burning it down or scalding the baby test.

I think the big differentiator for this one is the carrying capacity. They list 50kg instant/30kg sustained carrying capacity which is very impressive and I can't think of other humanoids with similar capability off the top of my head.

Could you post some links to the most impressive demo(s)? thanks

Reply on news.ycombinator.com