Humanoids will soon become great practice partners for tennis and other sports. These robots, trained by Galbot, are preparing to play with celebrities at the World Humanoid Robot Games in Beijing. They're fully autonomous.

OpenAI's work on humanoids is accelerating OpenAI "has 27 robotics jobs listed on its careers page, up from 11 in May, spanning hardware, software, data collection, and prototyping, according to Business Insider's review." "Publicly advertised base salaries range from $177,000 to half a million dollars a year, before equity."

Nobody is expecting humanoids to reach the home until at least 2030. Every timeline still puts full deployment some time away. @Figure_robot might have just disrupted this way of thinking. Index pretraining achieved a 56% success rate on unseen homes, compared to just 9% without the training. Cost was thought to be a large barrier for the end customers, just a few years ago, a humanoid would have been closer to $100k than $10k. Not only is that now not the case, @adcock_brett's lease approach brings them into a home for "something like $600 a month" How close are we to seeing a humanoid in every kitchen, living room & bedroom?

Training general-purpose humanoids will need unprecedented compute capacity, and Figure is stacking the chips. Figure, with cloud partner Nscale, is deploying up to 100,000 NVIDIA Vera Rubin GPUs. It's committing $3.5 billion initially, with plans to scale to $6+ billion. Deployment starts in the second half of 2027, at a data center in Barstow, Texas.

Goldman Sachs: Humanoids and memory Memory is expected to be up to 40% of semiconductor dollar content for humanoid robots (excluding GPU/CPUs). Goldman Sachs: Although content can vary, we assume an average system (DRAM) memory of about 128GB for a single humanoid unit, with a non-volatile (NAND) memory of about 1TB. Based on an assumed system memory of 128GB and a non-volatile memory capacity of approximately 1TB, we see a SAM range of between $250-$350 for DRAM and $350-$450 for NAND, for a total memory SAM of between $600-$800 or more per unit. Given the significant edge compute workloads of humanoid units, we expect that at least a portion of each system’s memory will rely on HBM, which is primarily supplied by Micron, Samsung, and SK Hynix where competition from Chinese memory makers (such as CXMT) may be limited in the near-term. Though whether humanoid manufacturers will decide to source HBM and traditional DRAM from separate vendors remains to be seen. Prices for DRAM and NAND memory can vary, though we utilize the rough market price for memory at the time of writing in our assumptions given that many LTAs have been struck at these levels. $DRAM

Liwei’s prediction: The humanoids robot trade will start fading next week. Capital rotates back to the real story: AI infrastructure buildout. If you’re still chasing robot stocks… it might be time to get out. The theme is simply too early. And after seeing the latest humanoid robot from UBTECH, I honestly had nightmares. Maybe it’s better if this technology develops a little more slowly. 😂