The cost structure of humanoid robots is undergoing a dramatic transformation! The proportion of semiconductors will surge to 24%, focusing on the three core areas of "brain, vision, and sensing."

The cost structure of humanoid robots is undergoing a dramatic transformation! The proportion of semiconductors will surge to 24%, focusing on the three core areas of "brain, vision, and sensing."

Following generative AI, humanoid robots are now being seen as the next disruptive technology hotspot, with a potential trillion-dollar market emerging.

According to Windchaser Trading Desk, Morgan Stanley stated in its recently released in-depth report—"Humanoid Robot Technology: Grasping the Future"—that humanoid robots, as the ultimate form of “Physical AI,” mark a crucial chapter in human history. The report notes that by 2050, the global humanoid robot market could reach $5 trillion, with 1 billion units deployed worldwide. Over this multi-decade growth, semiconductors will be the biggest variable and source of incremental value.

The report estimates that as the bill of materials (BOM) cost per robot falls from about $131,000 now to $23,000 in 2045, the share of semiconductor costs in the BOM will soar against the general trend, from the current single digits to 24%.

The decline in humanoid robot costs will dramatically shorten their investment return cycle, making them economically incomparable and driving mass adoption in factories, warehouses, farms, and other scenarios.

Rise of Physical AI: “Around One Humanoid Robot for Every 10 People” by 2050

The report highlights that Physical AI is the core of the broader AI theme—it “bridges the gap between AI and physical existence.” Unlike pure software AI assistants, embodied intelligence allows AI to understand and interact with the physical world.

Morgan Stanley forecasts that the humanoid robot market will see slow penetration before 2035, but will begin to accelerate significantly in the late 2030s. The key forecast data are striking:

  • By 2050, the global humanoid robot market will reach $5 trillion.
  • By then, global cumulative deployment will reach 1 billion units, equivalent to “about one humanoid robot per 10 people.”

This is a complex, fast-evolving ecosystem involving tech giants, startups, traditional industrial leaders, and world-class research labs. The report emphasizes, “Selective approaches are essential to seize the AI opportunity.”

Cost Revolution: The Path from $131,000 to $23,000

Cost is the key factor determining whether humanoid robots can be commercialized on a large scale. The report provides detailed BOM cost analysis, and its conclusions are encouraging.

The report estimates that "outside the Chinese supply chain, the current average BOM cost for one humanoid robot is about $131,000, but by 2045 this will drop to $23,000." This is far lower than the annual salary of most workers in developed markets. By 2050, with scaling and efficiency improvements, effective hourly cost could fall to $2.6, making them economically indispensable.

As costs fall, the composition of BOM value is also changing. The report predicts that from 2025 to 2030, BOM cost will increase by 15%, and by 2045, by another 40%. This is “mainly due to the increase in average selling price (ASP) for chips, since the computing intensity required per robot may rise, offsetting the general trend of declining semiconductor costs.”

Semiconductors: The Heart of the Value Chain

Within the value composition of humanoid robots, semiconductors are playing an increasingly important role. Morgan Stanley’s report shifts focus from the robot as a whole to the value of its internal semiconductors.

The report estimates that by 2045, the total addressable market (TAM) for humanoid robot semiconductors will reach $305 billion. This is equivalent to 49% of the global semiconductor TAM in 2024 ($627 billion). A key trend is that the share of semiconductors in the total BOM will “rise from the current 4-6% to 24%.”

This is mainly because, while costs for sensors and other components will fall as technology matures, the computing intensity required for each robot will keep increasing, boosting the value of core chips like AI processors.

Within semiconductors, value is highly concentrated. The report analyzes that “AI processors (chips) are the core components, currently accounting for 67% of total semiconductor costs, but by 2045 will reach close to 93%.” This highlights the absolute role of computing power in fulfilling the functions of humanoid robots.

Focus on Three Core Areas: “Brain, Vision, Sensing”

The report believes it’s not the robot integrators that are most valuable to the market, but the upstream companies providing core enabling technologies. It categorizes these opportunities in three key areas:

  1. Brain Technology: This is the core of humanoid robots, mainly including AI software and semiconductors, such as powerful GPUs, ASICs, or dedicated edge computing devices. These technologies are central to perception, decision-making, and communication capabilities. The report suggests that since the ultimate winner is not yet clear, investing in "leading advanced-process wafer foundries" is a relatively safe way to participate in this area.
  2. AI Vision: Enables Physical AI to "see" and interpret visual information. This requires “very high resolution cameras, high bandwidth and low latency, and advanced DSP chips.” Companies providing high-resolution camera solutions and cutting-edge digital image processing chips will be key beneficiaries.
  3. Sensing Technology: Analog chips are the core for perceiving the outside world, handling robots’ movement, sensing, and power, and are the foundation for hardware development. This includes sensors for temperature, pressure, or distance. The report specifically notes that “European analog chip companies are strategically well-positioned to benefit the most.”

“Top 25 Humanoid Robot Tech Companies”

According to Morgan Stanley, some companies are expected to stand out in the humanoid robot field and become core enterprises in the sector. The bank has listed the “Humanoid Tech 25,” comprising companies leading in technology, innovation, and market position.

Including Nvidia, Texas Instruments, Infineon, STMicroelectronics, Sony, and Samsung Electronics, among others.

Challenges & Risks: Three Gates to the Future

The report also points out the key challenges facing the industry:

  1. Technology & Cost: The development of specialized hardware, high cost management, and sensing/battery technology remain limiting factors.
  2. Energy Efficiency: Scaling AI faces two major bottlenecks: “(1) semiconductor fabs and (2) power plants.” Generative AI’s electricity demand is expected to exceed 75% of global data center power consumption in 2022 by 2027, and energy issues must be addressed urgently.
  3. Safety & Regulation: Ensuring the safety and reliability of human-robot interaction is paramount. Additionally, impacts on the labor market, data ownership, and sound regulatory frameworks are matters that require prudent handling.

~~~~~~~~~~~~~~~~~~~~~~~~

The above brilliant content is from Windchaser Trading Desk.

For more detailed interpretation, including real-time analysis and frontline research, please join the 【Windchaser Trading Desk · Annual Membership】

Risk Disclaimer and TermsThe market involves risk, investments should be made cautiously. This article does not constitute personal investment advice and does not take into account individual users' specific investment objectives, financial circumstances, or needs. Users should consider whether any opinions, perspectives, or conclusions in this article suit their specific situation. Investments made based on this, are at your own risk.