Humanoid data
Robotics companies are collecting vast amounts of data on human movement to train humanoid robots, using methods ranging from gig workers filming themselves to people controlling robotic arms remotely. The goal is to create robots that can easily slot into human workplaces and potentially replace humans.
Robotics companies are seeking vast amounts of data on human movement to train humanoid robots. They are using various methods, including paying gig workers to film themselves performing tasks and having people remotely control robotic arms. This data collection is driven by the success of large language models, which were trained on massive amounts of text data. Robotics companies believe that similar training data on human movement will help them build more capable humanoid robots. Training centers in China and gig workers in several countries are involved in this data collection. The future of work may involve physical laborers becoming data collectors, but it's unclear if this approach will yield technical breakthroughs or a profitable business.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.