Andrew's third point is Bandit's whole thesis.
If the robot brain goes open source, the edge moves to whoever can build hardware at scale, which is why our body will be built through the @machinedotfun
"WHAT BUBBLE?"
Andrew Kang, CEO and Co-Founder of our partner @RoboStrategy, says:
-Pushes back hard on robotics bubble talk. His math: every private robotics company combined is worth roughly $200B, while he sizes the opportunity as all physical labor, a market in the tens of trillions
-Says a robot that works 98% of the time isn't a good robot. Needing human help even 2% of the time is what separates real businesses from good demos
-Thinks if open source eats robot brains the way it's eating LLMs, the bottleneck shifts to hardware, and whoever can build robots at scale controls the value chain
-Points to labor shortages as the tailwind: the US is short hundreds of thousands of welders, and the people building data centers are now a bigger bottleneck than the GPUs
As the head of marketing I can't really tell what's going on here.
What I do know is it's in the simulator, we're testing shit (real), and Bandit is going take over the world.
When @jensenhuang asks for a somersault, you deliver. 🦆
See how AI agents on NVIDIA RTX Spark help bring @huggingface's Microduck from simulation to the real world.
Unitree Introducing: H2-X Kung Fu Legend Guardian🤖
Powerful performance to serve and guard you, with a maximum endurance of approximately 8 hours.
Please use the robot safely and in a friendly manner.
As a first update and 'Proof of Robot'
We’re currently testing different robot designs for Bandit.
This is a very early test. Nothing is final, and there’s much more to come.
We’re testing different bodies, movements and mechanical configurations before deciding what gets built.
Then the community gets a vote.
And it won’t stop with the design.
The community will help decide how Bandit moves, how he speaks, his special moves, where he goes and what he does next.