actuated<\/em> degrees of freedom.<\/p>\n\u201cWe have one of the most interesting hands in the world,\u201d Sleeper told me. \u201cTwenty-two degrees of actuated freedom. I think people don\u2019t really understand the differences here. People hear degrees of freedom, and they just take that as a given \u2026 they don\u2019t understand the variance. So 22 degrees of actuated freedom means you can actually actuate every degree of freedom that is available.\u201d<\/p>\n
Watch the episode here:<\/p>\n
When 1X announced Neo\u2019s hands at preorder launch in October 2025, the spec sheet said “22 degrees of freedom.” That number has become an industry benchmark, shorthand for human-level dexterity. But Sleeper draws a sharp distinction that spec sheets almost never make.<\/p>\n
Just because a robotic hand can move in a certain direction doesn\u2019t mean it can exert force. And just because it can close with power doesn\u2019t mean it can open with power.<\/p>\n
“A lot of people say 22 degrees of freedom, and it\u2019ll just be like these degrees of freedom are both able to move on the axes because they\u2019ve enabled that in the hardware, but you can\u2019t actually actuate them. So you can\u2019t actually do anything with that degree of freedom. Sometimes that makes it worse because then you can\u2019t control it.<\/p>\n
\u201cMost hands that are 22 degrees of actuated freedom — which I\u2019m not even sure there are any, probably the Chinese players in the hand space, but at least not for American humanoid players \u2014 you can actuate the hand closed, but you can\u2019t actually open it,” Sleeper adds. “It\u2019s usually just spring-loaded.\u201d<\/p>\n
In some cases, Sleeper says, uncontrolled passive joints actually make the hand worse, because the robot can\u2019t control what it can\u2019t actuate. And open-loop versus closed-loop actuation matters too. Most robot hands that can actively close their fingers spring open passively. There’s no feedback loop telling the system where the finger actually is when it opens. <\/p>\n
“You can actuate the hand closed, but you can’t actually open it,” Sleeper explained. “It’s usually just spring-loaded.” <\/p>\n
Without closed-loop control on every joint, you lose positional awareness, which means you also lose the ability to hold a resting, natural pose. And that\u2019s huge. Pretty much every humanoid robot you see has hands that look obviously unnatural. Obviously mechanical. Obviously unhuman, perhaps in a flat slab-like orientation with all fingers full extended, but certain not how most humans carry their hands, with a slight, natural-looking bend of the fingers.<\/p>\n
That matters psychologically, if we\u2019re going to have robots in our personal space. “There\u2019s nothing like shouting in your face, ‘I am an artificial being,\u2019 more than hands that are like this,” I told Sleeper, mimicking the rigid open-palm pose. He agreed, noting that similar uncanny-valley triggers exist throughout the robot\u2019s entire system, from the shape of the head to the way the feet move. 1X has worked hard, he says, to identify and systematically address each one.<\/p>\n
And the 22 degrees of actuation matters too, if we\u2019re going to have useful robots that can do significant chunks of the work we want them to do.<\/p>\n
That includes dressing themselves. And washing their own clothes.<\/p>\n
\u201cHere\u2019s one of the hard problems that we\u2019re actively iterating on,\u201d Sleeper says. \u201cWe have iterations of this suit coming through every week. We\u2019re iterating through multiple details on the suit to make it so Neo can take it off himself and clean it himself \u2026 if you had to do your humanoid\u2019s laundry, you\u2019d be pretty pissed.\u201d<\/p>\n
Yes. You\u2019d be your butler\u2019s butler.<\/p>\n
It\u2019s important to note that the hands on Neo right now that you might have seen in any existing videos are not the hands that will ship. 1X is working through multiple iterations, and the ones Sleeper showed me that were moving with incredible dexterity and speed are yet to be fitted to the robot. But these news hands, he said, will be on the shipping version. It\u2019s likely that Neo\u2019s AI might not yet be able to full take advantage of all that available hardware power. But, like any other smart product these days, that will change over time with software updates.<\/p>\n\nWhy tendon drive matters<\/h2>\n
1X is one of the only humanoid robot companies that are focusing on the home, perhaps primarily, in terms of sales. And there\u2019s a lot the company is building into Neo to make that work.<\/p>\n
The hands aren\u2019t an isolated engineering achievement: they\u2019re an expression of Neo\u2019s foundational architecture. 1X\u2019s proprietary tendon-drive system, which uses high-torque-density motors to pull flexible polymer tendons rather than driving joints directly with rigid gearboxes, is one of the things that make this level of dexterous control possible.<\/p>\n
That same architecture is what makes Neo quiet, operating at 22 decibels, quieter than a modern refrigerator. <\/p>\n
That\u2019s huge: most humanoids fit the \u201cclanker\u201d nickname you might have heard online. They\u2019re super noisy, which is not exactly what you might want in your home.<\/p>\n
“When it moves, you can’t hear it,” Sleeper told me. “It’s the faintest sound. I remember the first time we made a Neo video, we actually sampled the sound from Neo moving and lifted it in post when we were editing the video because we wanted people to hear a little bit of what a robot sounds like.”<\/p>\n
The contrast with the rest of the field is stark. <\/p>\n
Humanoid robots built on harmonic drives, a type of gear system used to transmit torque with very high precision, often used in industrial robots, can weigh 150 to 200 pounds. They tend to be loud, stiff, and have low back-drivability. (In other words, you can\u2019t easily brush their robot hands away.)<\/p>\n
1X has kept Neo at a relatively svelte 66 pounds and given him soft clothes: all to make the actual robot safe, and to make us feel psychologically safe around it.<\/p>\n
Insane levels of vertical integration<\/h3>\n
I pressed Sleeper on the price point, suggesting $20,000 was a loss leader number. While he didn\u2019t actually answer that specifically, he did point to a key factor in helping 1X keep costs low, even with manufacturing in the United States.<\/p>\n
The hands aren\u2019t outsourced. Neither are the actuators. According to 1X\u2019s own manufacturing blog, every hand begins with tendon material selection and in-house testing, moves through full tendon system assembly with custom electronics and 1X motors, and ends with proprietary polymer molding, including an advanced tactile sensing stack integrated throughout.<\/p>\n
Everything happens in Hayward, California, and that\u2019s the case with pretty much every component on the robot.<\/p>\n
“Because we\u2019re so vertically integrated, we make so many things in-house and we do all of our manufacturing here in the US, we\u2019re able to control our supply chain,” Sleeper said. \u201cWe also have 10-plus years now of innovation on motor design and manufacturing, tendon design and manufacturing, all this stuff. So we\u2019ve been able to drive the BOM [bill of materials] cost super low.\u201d<\/p>\n
That\u2019s theoretically what gets Neo to $20,000: a price point aggressive enough that the company\u2019s entire first-year production capacity of 10,000 units sold out within five days of preorders opening.<\/p>\n
What it\u2019s actually like to live with a humanoid robot<\/h3>\n
At this point, Sleeper is one of the few people on Earth who can answer that question from experience. And his account is more grounded than the glossy marketing might suggest.<\/p>\n
The value, he says, isn’t in the big dramatic tasks. Laundry \u2014 the end-to-end folding and putting away \u2014 is genuinely, profoundly hard. “If people tackle that, that’ll be on the list of full-self-driving-level timelines,” he said. That’s an honest admission from someone inside the company.<\/p>\n
Where Neo creates real value, Sleeper says, is in the accumulation of small things.<\/p>\n
The sweatshirt on the couch. The shoes left three feet from where they belong. The blanket not folded back into the basket, the dishes left out. These are the tasks that pile up in the background of daily life, the low-grade friction that drains your energy when you come home from work and before you\u2019ve sat down for the evening.<\/p>\n
“These little things actually really build up in your psyche,” Sleeper told me. \u201cYou walk into the house and you see it and you kind of ignore it at first because you’ve had a long day and you just want to get on to the thing you like to do in your home.\u201d<\/p>\n
So how long does it take acclimatize to having a humanoid robot in your home?<\/p>\n
Sleeper says it\u2019s minutes, not months. His Neo is certainly a novelty to guests, but they quickly get used to the robot. He points to how quickly humans adapted to talking to AI assistants and self-driving vehicles, things that would have sounded like science fiction just a few years ago. <\/p>\n
The home-first bet<\/h3>\n
I asked Sleeper about 1X\u2019s home-first bet.<\/p>\n
1X is nearly alone among American humanoid companies in explicitly targeting home consumers first. Most competitors like Figure AI, Agility Robotics and Apptronik are focused on industrial and logistics deployments. (Figure also has a significant home focus.)<\/p>\n
The conventional explanation is that factories are an easier sell: known tasks, controlled environments, business buyers who understand ROI.<\/p>\n
Sleeper offers a different reason. First, the home generates radically more diverse data than any factory floor. A robot handling pickup, retrieval, cleaning and social interaction in a real home encounters more variation in an afternoon than a warehouse robot sees in months, he suggests. That data diversity will make Neo\u2019s AI models smarter, faster.<\/p>\n
Second, the home is more forgiving of failure. In a factory, a 99.9% uptime requirement is default, expected. Miss it and you stop a production line. But in a living room, if Neo gets ketchup when you wanted a beer, you ask again. The home is where you earn the right to get good.<\/p>\n
(This suggests, by the way, that 1X will also target out-of-home markets.)<\/p>\n
Third, Sleeper makes a historical argument. Personal computers didn\u2019t become universal because enterprises adopted them: they took off because consumers did. He thinks the same dynamic applies to humanoids. The path from niche to mainstream runs through living rooms, not warehouses, in this way of thinking.<\/p>\n
Either way, Neo\u2019s sold out for now, and the company is working hard at adding production capacity in multiple U.S. locations.<\/p>\n
I know someone who\u2019s on the pre-order list now, and honestly, I\u2019m considering adding my name.<\/p>\n<\/section>\n<\/div>\n
\n
Source link <\/a><\/p>\n","protected":false},"excerpt":{"rendered":"1X’s humanoid robot Neo might just have the very best hands of any robot on (or soon to be on) the market, with 22 degrees of actuated freedom. 1X Before we started recording our recent podcast, 1X head of product and design Dar Sleeper showed me something actually amazing on his phone. It was humanoid<\/p>\n","protected":false},"author":1,"featured_media":14225,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[37],"tags":[],"class_list":["post-14224","post","type-post","status-publish","format-standard","has-post-thumbnail","category-brand-spotlights"],"_links":{"self":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts\/14224","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=14224"}],"version-history":[{"count":0,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts\/14224\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/media\/14225"}],"wp:attachment":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=14224"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=14224"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=14224"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}