Building an Artificial Mechanical Engineer:
All of the "AI CAD" demos you've seen have been essentially the same thing, a frontier lab LLM writing code that is translated by an existing geometry kernel into a 3d B-rep model.
I think this approach is fine, but generally lossy, and lacks true spatial understanding and geometric reasoning that humans naturally posses and apply when encoding their designs in a CAD system.
So I attempted to do something unique in the generative design space, building off of other generative 3d research, and trained my own (two tiny, 16M Parameter) AI models on a hyper specific set of data.
Introducing FieldGeo - a set of models that encode geometry, physics (FEA, Stress) and DFM into the same latent space, and reproduces (assisted with an additional design agent) geometries to meet the stated cases. Today, it's just for L shaped brackets.
This is some of the finest engineering Iβve seen in humanoid robotics so far. π¨
1X just showed a new tendon-driven hand for NEO, and this is exactly the kind of hardware humanoids needed.
Because the real test is not whether a robot can walk across a stage.
It is whether it can pick up a coin, hold a glass, plug in a cable, or use a tool without messing it up.
1Xβs new NEO hand is built for that:
β 25 degrees of freedom across the hand and wrist, so it can move closer to a human hand.
β Tendon-driven movement, so the hand stays soft, fast, and safe around people.
β Pressure, location, and slip sensing, so it can feel when something is about to fall.
β Up to 45N finger force, so it can grip real objects without becoming a
metal claw.
β IP68 waterproofing, so messy home tasks are actually possible.
β 2M+ cycle reliability, so this is being built for homes, not just lab videos.
This is where humanoids start getting interesting. Once the hand becomes good enough, the bottleneck shifts.
It is no longer βcan the robot physically do it?β
It becomes βdoes the AI have enough data to learn it?β
Humanoids were not waiting for a better chatbot.
They were waiting for hands that could actually touch the world.
@DrSonnyWhite Hey @DrSonnyWhite, what if we start extracting petawatts from the local spacetime fabric, would that affect it in any way? Is it a truly inexhaustible source of energy? Could mega scale extraction effects be observed remotely? Could be an interesting SETI target if so..
We are back. After one year of quiet building.
Introducing GENE-26.5, our first robotic brain that takes a major step toward human-level capability.
For years, robotics has struggled to learn from the worldβs largest and valuable data source: Humans.
Solving it means rethinking the whole stack from the ground up:
- A robotics-native foundation model.
- A 1:1 human-like robotic hand.
- A noninvasive data collection glove for motion, force, and touch.
- A simulator that turns weeks of experiments into minutes.
GENE-26.5 is trained across language, vision, proprioception, tactile, and action. We designed a set of tasks to test how far we can go with this new paradigm.
Fully autonomous, 1x speed, one model, same weights. (Enjoy with sound on)
We are approaching the endgame for robotics.
And this is just a beginning.
@ExtrovertedDev@DaveShapi They cannot create human-made art. They cannot be human and there will, for ages to come, be people who insist on having certain interactions exclusively with humans, or to demand certain objects made by human hands for the sake of it. Just look at the hand-made market today.
@mlejva@braelyn_ai The whole pre LLM internet is now like this. Think about it often. Any old blog has suddenly regained if not multiplied many times over its original value.
@Vexarian@G0ffThew@waitbutwhy > You think people are morally obligated to save others from the consequences of their own actions.
That's right, I do feel like I have that responsibility.