In factories around the world, dedicated robotic arms have long dominated industrial production—powerful, precise, and cost-efficient. They excel at repetitive, single-task jobs, from welding car frames to packaging goods, and have reshaped modern manufacturing in profound ways. At first glance, investing in humanoid robots might seem illogical: why pour resources into a complex, human-shaped machine when specialized robots already perform so well? Yet I firmly believe humanoid robots are not just a technological experiment, but the next transformative leap for humanity—and that is why I bet big on them.
The core value of humanoid robots lies in their ability to adapt to the human-centered world we have built for ourselves. Unlike fixed industrial robots, which require entirely customized environments and workflows, humanoid robots are designed to navigate spaces made for humans: staircases, doorways, offices, homes, and crowded public areas. They can use standard tools, operate everyday appliances, and perform a wide range of unstructured tasks without the need to rebuild our surroundings. A factory robotic arm can only do one job; a humanoid robot can clean a house, assist the elderly, restock shelves in a store, and help in a hospital—all without reconfiguration. This general-purpose versatility is something no specialized machine can ever match.
From first principles, a generalized machine makes perfect sense. Specialized machines and robotic arms will still exist and serve their roles in the future, but general-purpose robots will take on a much broader scope of work. We can use the iPhone as a perfect analogy: we used to have separate devices like typewriters, cameras, DVD players, and music players, but the iPhone combined all of them into one handheld device. The same transformation will happen with humanoid robots. They will become the all‑in‑one general-purpose machine for the physical world.
Another critical point people often miss is the economy of scale. Today, dedicated robotic arms may seem simpler and cheaper, but a device’s cost is not determined by how simple it is—it is mostly determined by how many units are produced each year. Right now, the supply chain for humanoid robots is still immature, and that takes time to develop.
But here is the deeper logic: if there is no single, leading, standardized shape for a general-purpose machine, it is almost impossible to reach real economy of scale.
Look at what happened with cars about 100 years ago. Once the industry reached a consensus on the basic form of a car—four wheels, engine, cabin, a standard way to move people from point A to point B—production volume exploded, costs dropped dramatically, and entire industries were transformed.
Today, there are hundreds or even thousands of different types of robotic arms and specialized machines, many with overlapping functions. But if we do not build them in human shape, what shape should we agree on? It is extremely difficult for the world to reach consensus on any other form factor.
The human body is not perfect for every single task—for heavy mining or large-scale industrial work, we will still need giant, specialized machines. But for most daily, general-purpose tasks, the human form is surprisingly smart and versatile.
If the world can align around the humanoid form as the standard general-purpose robot, it will create a gravity center for human intelligence and innovation, just as Steve Jobs once described. Over the past 20 years, so much human talent and technology converged into one handheld device—the smartphone—and that changed the world. The same will be true for humanoid robots.
A unified, standardized human shape will allow the entire industry to focus its research, supply chain, manufacturing, and software development on one platform. This will push productivity to new heights and drive production costs down exponentially. When no better alternative exists, the human-shaped machine is simply the most logical path forward.
Humanoid robots also solve the most pressing labor challenges of our time. As populations age in many countries and labor shortages grow in sectors like healthcare, eldercare, logistics, and household services, we need machines that can work alongside humans, not replace them in isolated factories. Humanoid robots can take on physically demanding, dangerous, or monotonous tasks, freeing humans to focus on creative, emotional, and strategic work. They do not compete with industrial robots—they complement them, extending automation from controlled factory floors to the unpredictable, dynamic spaces of daily life.
Beyond practicality, humanoid robots represent the future of general artificial intelligence and adaptive technology. Many people misunderstand where the real bottleneck is. The challenge is not hardware; we already have capable actuators, motors, and structural designs. The real limitation is intelligence: humanoid robots still struggle to perceive, understand, and interact reliably with the real 3D world. This means the AI software is far more important than the mechanical body.
Elon Musk once said that we are waiting for the ChatGPT moment for humanoid robots, and I firmly believe this to be true. Before that AI breakthrough arrives, humanoid robots will find it hard to enter a truly mass market and drive a real revolution for human society. However, this does not mean there are no viable niche markets today. For example, humanoid robots with a natural appearance and basic conversational ability can already serve as receptionists or customer service roles, providing steady cash flow to support the industry’s growth.
Companies like Tesla, with its strong AI capabilities and massive computing infrastructure such as xAI and Dojo, are in an advantageous position to lead this transition. They have the data, the compute, and the long-term vision to build the advanced AI that humanoid robots desperately need. This is not just about building a robot—it is about building a machine that can truly understand the world.
Critics may argue that humanoid robots are too expensive or complex compared to specialized robots. But history shows that transformative technologies start out imperfect and costly before becoming accessible and indispensable. Personal computers were once niche and expensive, yet they replaced typewriters and changed the world because they were flexible and general-purpose. Similarly, humanoid robots will evolve from experimental machines to essential tools in homes, workplaces, and communities.
Dedicated robotic arms have revolutionized manufacturing, but they are limited to one role in one place. Humanoid robots are built for the entire human world—versatile, standardized, scalable, and capable of integrating into every aspect of our lives. They solve real, urgent problems and push the frontiers of what technology can do. We just need to be patient and wait for that ChatGPT moment.
Until then, the industry will grow step by step, supported by niche applications. But once general intelligence arrives, combined with the power of economy of scale, humanoid robots will redefine how we live and work.
For all these reasons, I am not just optimistic about humanoid robots—I bet big on them.
欢迎所有朋友加我微信：happypeter1983