Comment In early October, Tesla demo’d Optimus, its humanoid robot. The company’s previous demo had involved marching a human out in a robot-like bodysuit, so when Optimus walked slowly around the stage, it was met with delight from the cheering crowd. Despite the show’s futuristic framing, robotics experts were mostly underwhelmed by the reveal. Optimus’ clunky attempts at something like a dance seemed less advanced than other humanoid robots, such as Honda’s ASIMO, which played soccer with former President Barack Obama back in 2014. Tesla engineers boasted that Optimus’ hand had as many as 11 degrees of freedom (that’s to say, all the ways in which robotic parts can bend). In comparison, a robotic hand designed by a Japanese engineer back in 1963 had 27. Despite its obviously limited capabilities, Optimus still triggered a familiar robots-will-someday-rebel-against-their-creators anxiety. What is it about Optimus that makes us feel threatened? My research on the development of Japanese robotics reveals that our feelings about robots are less about the general idea of artificiality, as many critics suggest, and more about the fact that robots are proxies for real human beings. People’s feelings toward robots often mirror their feelings toward the kind of human worker they imagine the robot is meant to replace. Watching the long video of the demo, I was astounded to hear Elon Musk and his engineers make statements that echo what Japanese engineers said back in the ’80s (and even earlier): the need for “biologically inspired design” to create a multipurpose machine, the promise that robotic workers will liberate us from the dread of work and bring overall happiness and prosperity in 10 to 15 years, and the desire to have a robotic companion, to name a few. The experience of Japanese engineers who tried to do (some 40 years ago) what Tesla engineers are trying to do now is revelatory, both because it shows that the task is much more difficult than it seems, and also because it teaches us to identify hidden assumptions and biases that go into designing a humanoid robot. Advertisement Advertisement In the case of Japanese androids, service robots are often diminutive and feminized (with a tiny chin and nose, no jaw line, and a high-pitched vocal range). The design of their bodies offers visual hints to the kind of humans they are meant to replace. The subtle apron visible on some Japanese service robots, for example, is associated with the caring “auntie.” These robots are expressly designed to give the user a sense of family sympathy, of being taken care of, of being important. In contrast, robotic “receptionists” are modeled on hypersexualized young women, a design that communicates that the labor expected from them is not strictly clerical. (Sexbots are another story—and the reason that Japanese engineers have worked so hard to develop super realistic artificial skin.) Advertisement That brings us back to Optimus. Musk stated that it was possible “to put all kind of costumes on the robots.” So, what does Optimus’ design—as opposed to the drawn prototype or the man in the suit who modeled it—reveal about the kind of person it (or, as the designers would have us believe, “he”) is dressed as? First, he is undeniably male, with a tall stature, exaggerated quads, and broad shoulders. Optimus’ tiny head telegraphs that he is not a thinker. The blank slate where one might expect a face assures us that his emotions do not matter. The design tells us that the human worker Optimus is modeled on is valued for his manual labor alone. He does not think. He labors. He does what is asked. Advertisement It’s also worth noting that, in the demo, Optimus’ face and hands are black. Does that mean anything? Perhaps. Many American users may (sometimes unconsciously) still make a connection between male manual labor and Blackness. I once read the transcript of a conversation between Japanese engineers, who, in 1973, reflected on their belief that America sought to design robots in order to replace its Black labor force. I have no doubt that Japanese engineers were reading American magazines, where robots were imagined to take over jobs often associated with Black Americans—hard manual labor, manufacturing, and trash collection. And some American articles are even more explicit about their intentions. My colleague Jason Resnikoff cites a 1957 article from Mechanix Illustrated that promises readers that they “will own ‘slaves’ by 1965.” That is, “robotic” slaves—robotic butlers, cooks, drivers, cops, secretaries, and security guards. Optimus could thus be interpreted as a representation for both oppressive labor structures and the racialization and subsequent devaluation of different kinds of labor. The fear Optimus evokes among some Americans, perhaps, is a fear of rebellion against those structures. Advertisement