• This story originally appeared in The Algorithm, our weekly newsletter on AI. • To get stories like this in your inbox first,sign up here. • In January, Nvidia’s Jensen Huang, the head of the world’s most valuable company, proclaimed that we are entering the era ofphysicalAI, when artificial intelligence will move beyond language and chatbots into physically capable machines. • (He also said the same thing the year before, by the way.) The implication-fueled by new demonstrations of humanoid robots putting away dishes or assembling cars-is that mimicking human limbs with single-purpose robot arms is the old way of automation. • The new way is to replicate the way humans think, learn, and adapt while they work. • The problem is that the lack of transparency about the human labor involved in training and operating such robots leaves the public both misunderstanding what robots can actually do and failing to see the strange new forms of work forming around them.

Article Summaries:

  • Nvidia’s CEO Jensen Huang has touted a shift to “physical AI,” where robots learn to act like humans rather than perform single‑purpose tasks. The article argues that this new era masks the extensive human labor required to train and operate humanoid robots. Examples include a Shanghai worker wearing a VR headset and exoskeleton to repeatedly open microwave doors for data collection, and U.S. firm Figure’s planned partnership with Brookfield to harvest household data. Other companies, such as 1X, rely on tele‑operators who remotely control robots for chores, raising privacy and wage‑arbitrage concerns. The piece warns that the public’s understanding of robot capabilities is obscured by these hidden labor practices.
  • The article highlights how the growing field of humanoid robots relies on hidden human labor for training and operation. Workers in places like Shanghai and North America are now tasked with repetitive demonstrations-wearing VR headsets, exoskeletons, or movement‑tracking sensors-to generate the data that teaches robots to perform household chores. Companies such as Figure and 1X plan to collect massive amounts of real‑world data or employ remote tele‑operators to guide robots when they encounter difficult tasks. This practice raises concerns about transparency, privacy, and the potential for wage‑arbitrage that mirrors gig‑work dynamics while enabling physically capable machines.

Sources: