The editorial argues the $1.6B implied valuation is absurd given Figure AI's $39.5B, 1X's $10B, and Physical Intelligence's $5.6B valuations. Boston Dynamics has the most mature humanoid hardware and a 30-year head start on locomotion, yet just changed hands at a price that wouldn't crack a Series B in today's robotics market.
Notes that this is Boston Dynamics' fourth owner, with each handoff (MIT → Google → SoftBank → Hyundai) happening because the prior owner couldn't turn locomotion demos into revenue. The IP and team aren't the problem — Atlas's electric transition and Spot's enterprise traction prove the tech works; capital markets just price the company on a different curve than narrative-driven competitors like Figure and 1X.
By surfacing the story, the submitter highlights that Hyundai going from 80% to 100% ownership — rather than spinning Boston Dynamics off or letting SoftBank ride along — points to a strategic intent to fully integrate humanoid and quadruped robotics into Hyundai's manufacturing and mobility roadmap. The 763-point score reflects developer community interest in what a fully Hyundai-owned Boston Dynamics builds next.
SoftBank has sold its remaining 20% stake in Boston Dynamics to Hyundai Motor Group for approximately $325 million, giving the Korean automaker 100% ownership of the company best known for Atlas, Spot, and a decade of viral parkour videos. The deal closes a transition that began in 2020, when Hyundai acquired a controlling 80% stake from SoftBank in a transaction valued at roughly $1.1 billion.
The implied valuation — about $1.6 billion fully diluted — is the part that should make every robotics investor squint. Figure AI most recently raised at a reported $39.5 billion valuation. 1X is reportedly raising at $10 billion. Physical Intelligence closed a $400M round at $5.6B. Boston Dynamics, with arguably the most mature humanoid hardware on the planet and a 30-year head start on locomotion research, just changed hands at a price that wouldn't crack a Series B in 2026's robotics market.
This is Boston Dynamics' fourth owner. It spun out of MIT in 1992, was acquired by Google's Replicant moonshot group in 2013, sold to SoftBank in 2017 after Alphabet decided robotics wasn't a business it wanted to be in, and partially sold to Hyundai in 2020. Each handoff happened because the prior owner couldn't figure out how to turn locomotion demos into revenue.
The story here isn't 'SoftBank exits another investment' — Masayoshi Son does that on a Tuesday. The story is that Boston Dynamics' valuation has been flat-to-down for a decade while every other humanoid robotics company has gone vertical. That's not a comment on the technology. Atlas's hydraulic-to-electric transition last year produced the most fluid bipedal motion ever publicly demonstrated. Spot is a real product with real customers in oil & gas, construction, and utilities. The IP and the team are not the problem.
The problem is that capital markets in 2026 are pricing humanoid robotics on a different curve than they were in 2020. The Figure/1X/Tesla Optimus narrative — humanoids as general-purpose labor that replaces warehouse workers within 36 months — has pulled valuations into the same speculative bracket as foundation-model labs. Boston Dynamics, owned by a Korean car company and run as a strategic R&D unit rather than a venture-backed moonshot, doesn't get to surf that wave. It can't raise a $1B round at a $20B markup because its owner isn't selling shares.
The inverse take: Hyundai is paying a relative bargain for capability that competitors are spending years and billions to replicate. SoftBank, for its part, has been raising cash for OpenAI and Stargate and likely needed liquid capital more than a minority stake in a hardware company with thin margins. The $325M exit looks small next to what Boston Dynamics could plausibly fetch in a hot SPAC or IPO market — but SoftBank has been burned enough by waiting for the perfect window (see: Arm's IPO timing) that taking the cash now is a defensible call.
Community reaction on Hacker News has been pointed. The top comment thread is essentially: 'Google sold this for cheap, SoftBank sold this for cheap, and now we'll find out whether vertical integration with a car factory is finally the right home or just the next stop.' A counter-thread argues this is the best possible outcome — Hyundai actually builds physical things at scale, has assembly lines that need humanoid labor, and has been integrating Atlas-derived tech into its factories for two years already.
If you're building in robotics, simulation, or embodied AI, the signal here is about deployment surface, not technology. Boston Dynamics is now structurally a captive R&D arm of a car manufacturer with 5 million units of annual production capacity and a real, immediate need for humanoid factory labor. That's a more concentrated, more boring, and probably more lucrative business than competing for the open-ended 'general-purpose humanoid' market. Expect Atlas's roadmap to bend hard toward automotive assembly tasks — wire harness routing, fastener insertion, quality inspection — and away from showy parkour demos.
For anyone working on robot foundation models (the RT-2, π-0, OpenVLA crowd), this consolidation matters because the largest non-Tesla source of real-world humanoid telemetry data just moved fully inside one company's walls. Hyundai will have every incentive to keep that data proprietary, which constrains the open ecosystem the way Waymo's data lock has shaped the AV research community. If you've been hoping Boston Dynamics would open-source meaningful chunks of its locomotion stack, lower your expectations.
The second-order effect: Hyundai now competes directly with Tesla on the 'car company that also makes humanoids' axis, and unlike Tesla, Hyundai actually has Boston Dynamics-grade balance and manipulation. Optimus has the demo-narrative advantage; Atlas has the engineering. Watch for Hyundai to start publishing factory-deployment metrics within 12 months as a pointed contrast to Optimus's perpetually-pushed-out timeline.
The interesting question isn't whether Hyundai can run Boston Dynamics — it's already been running it for five years. The question is whether being a captive R&D arm of a manufacturing giant is the *correct* business model for humanoid robotics, and whether the Figure/1X/Optimus valuations look prescient or insane in 36 months. If humanoids actually become factory labor at scale, Hyundai just bought the technology stack for what will look like a rounding error. If humanoids remain demo-ware for another decade, SoftBank gets to claim it timed the exit. Either way, the era of Boston Dynamics as an independent, ownership-shuffled research lab is over. It's a Hyundai subsidiary now, and the next Atlas video will probably be filmed on a production line.
I don't understand why they would implement humanoid robots instead of purpose-built robots. The human form is not the most optimal way to do most tasks, especially as it relates to manufacturing. Robots don't need to look like humans, they need to be useful. Seems like putting in an awful
I don't think this is solely tied to car manufacturing automation. Even though Hyundai Motor Group is acquiring them, I would imagine they'd be well-positioned to commercialize general-purpose robotics and not just for car manufacturing, if Tesla is anything to go by.I do think this might
Wait, haven’t they already owned them for years? Edit: right, they’re just buying the remaining 9%.
Hyundai bumped their ownership up to 100%, and took the opportunity to reset expectations about when Atlas would be working in Hyundai factories.While Atlas is the best humanoid robot so far, it still isn't useful in a car factory that's fully equipped with the latest factory robots that a
Top 10 dev stories every morning at 8am UTC. AI-curated. Retro terminal HTML email.
Back in December 2020, Hyundai purchased an 80% controlling interest in Boston Dynamics from SoftBank for $880 million, part of a transaction that valued the robotics company at $1.1 billion. That agreement included a put option allowing SoftBank to sell its remaining stake to Hyundai at a later dat