Schaeffler said on 21 August it has finished validation testing on formed strain wave gearboxes built specifically for humanoid robots and will manufacture them at scale from 2027.
What it means for your work
Humanoid pricing has been held up by hand-built actuators; an established supplier committing a production line to them is what moves the cost curve.
Our read on the items that move something. The reporting is everyone's; this part is ours.
1
The retrofit-onto-fleet-you-already-own pattern keeps winning the money in construction robotics — the productivity claim is the vendor’s, but the deployment model means a contractor can pilot it without a capital purchase.Gravis Robotics raises $200M from SoftBank to retrofit excavators into autonomous machinesThe Robotics & Physical AI beat · The Robot Report
2
Humanoid programmes are gated on actuator supply rather than on demos, so a tier-one supplier committing to volume is the real constraint moving.Schaeffler will mass-produce humanoid-robot gearboxes from 2027The Robotics & Physical AI beat · The Robot Report
3
Driving at fleet scale turns out to be an inference-hardware problem as much as a model problem, and this is a rare public number for it.Waymo names the silicon behind its robotaxis — a purpose-built 5nm chip above 1,000 TOPSThe Robotics & Physical AI beat · Robotics & Automation News
Following opens with early access — sign up for early access to register your interest; it does not switch following on today. Everything you follow will land on one page, in your account menu.
All coverage
Everything Intel has read, newest first. Each title opens at its original publisher.
Construction is becoming the proving ground where robotics research meets an acute real-world labor gap — with university pipelines now funded to industrialize it.
Whole-body control plus few-hour adaptation to new hardware is what turns humanoid demos into deployable platforms — the robotics stack is consolidating around foundation models.
Physical AI has no internet-sized corpus to scrape, so the unit economics of annotation — not model architecture — is what decides how fast robots get good.
Brain-computer interfaces have mostly lived in research demos and rehab clinics; a sub-200ms decode loop that plugs into off-the-shelf robots is what turns "thought control" from a lab trick into a control scheme other robotics teams can actually build on.
A robot policy that works in a demo can fail in the world — shared, large-scale evaluation in simulation is the missing benchmark layer physical AI needs before it is trusted off the lab floor.
When a frontier lab puts its vision-language-action models directly in startups' hands, embodied AI stops being a lab demo and starts becoming an ecosystem — the same pattern that scaled language-model apps.