Robotics Researchers Doubt Humanoid Robots Will Arrive Soon Despite Musk’s 2027 Prediction

The article, published in collaboration with the nonprofit research foundation Aventine, describes Optimus as a white robot with a black head and torso that has appeared in videos dancing, passing popcorn, putting trash in a bin, vacuuming, and pressing a microwave button. It has also been shown falling backward while handing out water bottles and struggling to iron a shirt. For now, the robot is most often seen handing out food and drinks at Tesla events.

Musk, Tesla’s CEO, has told shareholders that Optimus will be “not just Tesla’s biggest product ever, but probably the biggest product ever,” headed first to factory floors and later to homes. In July, he said it “will have human and then superhuman dexterity.” At the World Economic Forum’s annual meeting in Davos, Switzerland, in January, he predicted the robots could be on sale to the public by the end of 2027. He has argued that Optimus robots could automate almost all human labor—from hauling sheet metal to folding laundry—for as little as $20,000 each.

Other technology leaders have made similarly expansive predictions. Marc Andreessen, cofounder and general partner of Andreessen Horowitz, has said robotics could become the “biggest industry in the history of the planet.” Nvidia CEO Jensen Huang said in January that humanoid robots would match human-level ability this year. Morgan Stanley estimates that the number of robots that “resemble and act like humans” is likely to reach nearly 1 billion by 2050, creating a market worth more than $5 trillion.

Those proclamations are largely fueled by the idea that the same AI advances behind OpenAI’s ChatGPT and Anthropic’s Claude will let a new generation of robots imitate human movement the way chatbots imitate human language. But many robotics researchers are skeptical, arguing that such assumptions minimize the challenge of using an intelligence built on language and images to master the infinite variability of the physical world.

“None of those companies [building humanoid robots]—absolutely none of them—has any idea how to make those robots smart enough to be useful,” Yann LeCun, often described as one of the godfathers of AI, said at another event during the January Davos conference. Researchers also say it is misleading to conflate humanoid robots made to resemble people with generalist machines able to learn and perform multiple tasks.

“It’s very easy to make a robot that looks like a person,” said Jonathan Hurst, cofounder and chief robot officer of Agility Robotics and a professor of robotics at Oregon State University. “It is dramatically more difficult to make a machine that moves or behaves dynamically or physically like a person.”

The tensions over whether all-purpose humanoid robots are just around the corner or nowhere in sight—and whether current forms of AI are all that is needed to perfect them—are playing out in robotics labs across the country, where hype over timelines is obscuring painstaking but meaningful progress, the article says. A decade or so ago, a series of breakthroughs led to the generative AI revolution. Roboticists disagree on when it will happen, but believe an equally transformative revolution is possible in robotics, one that would give machines physical intuition and fluidity that has long been out of reach. The question is whether the same methods and tools that fueled AI advances are enough, or whether an entirely new path is required.

At Google DeepMind, researchers are testing one of today’s most advanced robot brains on a piece of equipment called ALOHA 2, short for “A Low-cost Open-source Hardware System for Bimanual Teleoperation.” Roboticists have long clashed over whether a humanlike form is necessary for generalist robots. Proponents argue it helps them fit into the world as it exists; detractors say it is not worth the trouble. ALOHA 2 reflects the second view. It consists of a pair of arms, some grippers, and a couple of cameras.

Despite its simplicity, ALOHA 2 is a workhorse for Google DeepMind researchers, who use it to test their most advanced AI for robotics system, Gemini Robotics, in various labs. When controlled by Gemini Robotics, ALOHA 2 becomes more of a generalist robot, able to perform tasks based on examples it has been trained on, the article says. One example described is packing a lunchbox.