The Humanoid Market

What is a humanoid?

In January 2025, sensational news proclaiming the arrival of the humanoid era spread across the globe. NVIDIA’s CEO, Jensen Huang, introduced the concept of “Physical AI”—AI operating in the physical world—and remarked, “ChatGPT moment for general robotics is just around the corner."¹

So, what exactly is a humanoid? While various definitions exist, following NVIDIA’s perspective, a humanoid is defined as “a general-purpose, bipedal robot designed in a human-like form, capable of learning and performing a wide range of tasks.”²

Compared to traditional industrial robots, the primary distinguishing feature of humanoids is their human-like appearance. Furthermore, equipped with advanced AI, humanoids can dynamically perceive their environment and autonomously perform assigned tasks by leveraging skills acquired through training —moving and acting much like humans.

Why humanoids "now"?

There are two key reasons why humanoids are attracting significant attention right now.
 

1. Demand Side

In recent years, a growing shortage of labor—especially in developed countries—has led to increasing market interest in humanoids. In the United States in particular, where rising labor costs due to inflation are putting pressure on corporate profits, expectations are rising for humanoid robots as an effective solution for reducing costs in the medium to long term. Additionally, there is strong interest from stakeholders in deploying humanoids in hazardous environments—such as defense, disaster response, and even space exploration—where human involvement carries significant safety risks. These factors are fueling expectations for humanoid development.
 

2. Technology Side

Recently, rapid advancements in software—especially AI—have reached a level sufficient to enable practical humanoids, accelerating development in this field. Advancements in synthetic data generation and simulation tools, such as digital twins, have also accelerated, supporting more effective AI training. On the hardware side, key components such as sensors, batteries, and actuators have become smaller and more precise, making it possible to develop humanoids that can move and perform tasks in a human-like manner.

Thus, the period between 2025 and 2030 is when (1) growing demand and (2) technological advances will intersect, and it is expected that the widespread adoption of humanoids will accelerate during this period.

Humanoid Market Overview

Various forecasts have been published regarding the size of the humanoid market, but one of the most sensational recent projections comes from Morgan Stanley: by 2050, one billion units could be deployed globally, with final product sales alone reaching a scale of USD 4.7 trillion.³ This is a shocking figure, nearly double the combined revenue of the world's top 20 OEMs as of 2024.

Another forecast for the nearer future is that the global market will reach USD 29 billion by 2033. ⁴ This figure corresponds to an impressive compound annual growth rate (CAGR) of 34.6% from 2024.

According to GlobalX’s unit-based forecasts, sales volumes are expected to begin gradually increasing around 2026, though the pace will be moderate at first. The market is anticipated to reach an inflection point in adoption around 2030, signaling a period of accelerated growth.⁵

Meanwhile, experts from U.S. humanoid startups present a slightly different perspective. They consistently state that, at least for sectors leading in use case development—such as logistics and manufacturing—the widespread adoption of humanoids is likely to begin around 2027, about two years from now.

According to these experts, this projection is based on the expectation that some of the most critical hurdles to widespread market adoption will be overcome within the next two years. Once these challenges are addressed, they anticipate that the market will be poised for rapid growth, particularly in logistics and manufacturing.

Challenges in humanoid development and efforts to resolve them

Humanoid development faces numerous challenges, ranging from technical issues to safety and development and production costs.
 

Technical Challenges

On the technical side, controlling AI in the physical world is inherently difficult. In addition to the need for vast amounts of training data, there is also the practical requirement to secure physical spaces where robots can be safely trained and tested. This poses a significant constraint.

Assuming humanoids are to be used as human replacements, manufacturing and logistics companies expect them to be able to operate a typical work shift without recharging—that is, they must have battery performance and power efficiency sufficient to operate continuously for 8–10 hours. Currently, this standard has not yet been achieved, though some companies are attempting to address the issue through battery swapping solutions.
 

Safety Considerations

Perhaps even more critical is the issue of safety. When humanoids are deployed in environments where humans are present, it is absolutely essential that they do not pose any risk of harm. However, safety standards for humanoid robots have yet to be established in most countries, making this a major hurdle for widespread adoption.
 

Rapid Progress Toward Overcoming Key Challenges

Efforts to resolve these issues are currently progressing rapidly. For example, the International Organization for Standardization (ISO) is already in the process of establishing safety standards for humanoids. ⁶

According to U.S. humanoid experts, startups are making significant breakthroughs in battery technology, with the potential to achieve the performance required for 8–10 hours of continuous operation in the near future.

Performance improvements are accelerating. The market will only embrace humanoids at scale when their capabilities are proven to match or exceed those of human workers in key use cases. Currently, leading humanoid models in the U.S. reportedly achieve around 70% of human performance in certain logistics tasks. Startups are striving to reach 100% or higher in the near future. If these advancements continue as expected, the next two years could mark the beginning of explosive adoption.
 

Production Costs as a Key Constraint on Humanoid Adoption

Production costs are another critical factor that could limit the pace of humanoid adoption. Currently, U.S.-developed humanoids designed for logistics and manufacturing are estimated to cost at least USD 150,000–200,000 per unit. However, experts believe that hardware costs will decrease rapidly as mass production scales up.

In the long term, some specialists suggest that Elon Musk’s prediction—a retail price of around USD 20,000–25,000 per unit⁷—may not be out of reach. This significant reduction in cost could be a game-changer, making humanoids far more accessible for widespread deployment in industrial settings.

The Humanoid Value Chain

The value chain for humanoids can be broadly divided into four main segments: ⁸

• Brain: This segment includes software such as AI foundation models, simulation technologies for training, semiconductors, memory, etc.
• Body: This includes the actuators to convert energy into mechanical movement, various sensors to capture the external environment and objects, batteries, and metal and plastic materials used to create the exoskeleton and the robot's surface.
• Integrators: This includes players who integrate brains and body parts to create the final humanoid product.
• Service development: To deploy humanoids across various applications, integration with current business operations and automation systems is essential. This function will be handled by systems integrators and consulting firms, among others. Currently, integrators often supply humanoids directly to user companies as Robotics as a Service (RaaS), resulting in limited involvement from service development providers. However, this category is expected to become increasingly important as the development and adoption of humanoids progress. In the future, additional services—such as maintenance, insurance, and financial services—are expected to be incorporated into the humanoid value chain, similar to other machinery and vehicles.
  

Key players

The humanoid robotics sector is shaped by several distinct types of key players, each with their own motivations and strategies:

1) Large global corporations: Some global enterprises aim to leverage humanoids for medium-term operational excellence, while others see the sector as a high-potential market for future growth and are seeking new entry opportunities. Hyundai’s acquisition of Boston Dynamics reflects their strategies for both operational excellence and market expansion. ⁹

2) National governments: Governments play a major role in supporting humanoid development. Humanoid development requires a large amount of capital, making it extremely important for companies to obtain strong support not only from private investors but also from the public sector. Currently, the two overwhelming leaders in humanoid development are undoubtedly the United States and China, with China in particular supporting humanoid development as a national policy, and serving as a driving force behind rapid development progress for humanoids.
Although not widely publicized, there are indications that governments are exploring humanoid applications in defense, especially as geopolitical tensions rise—adding another layer to government involvement.

3) Big Tech: Big Tech firms are indispensable in this market, positioning humanoids as a major business opportunity within their AI, software, and hardware portfolios. Typical examples include NVIDIA, which is leading the way with its foundation model called GR00T, and Tesla, which is accelerating the development of its own humanoid robot as an integrator. Amazon is collaborating with leading startups (integrators) to test humanoids in its logistics operations. ¹⁰

4) Startups: Startups worldwide are racing to establish their humanoid models as industry standards, supported by investments from governments, Big Tech, and private investors. Leading startups are concentrated almost entirely in the United States and China, and are developing and verifying use cases in collaboration with major companies and Big Tech in each country.

Progress in Humanoid Use Case Development

Currently, the development of use cases for humanoids is progressing particularly in the manufacturing and logistics fields. In fact, commercialization of humanoids has already begun in the logistics field, albeit in a small area, and humanoids are being used primarily for simple repetitive tasks that can be completed within warehouses, such as picking and carrying items.

In manufacturing, especially on automotive production lines, pilot projects are underway to test humanoid integration. Since their production lines are designed for human workers, the humanoid form factor offers a distinct advantage, allowing robots to fit seamlessly into existing workflows.

On the other hand, it is generally expected that practical implementation in sectors other than these two will take several more years.

For example, healthcare and nursing care applications, which face serious labor shortages, are one area where humanoids are highly anticipated. However, because these applications involve direct contact and communication with people, widespread adoption will require the establishment of sufficient safety standards as a prerequisite. Moreover, the search for specific ‘killer applications’ is ongoing, and widespread adoption is expected to take time. Retail is another industry facing severe labor shortages. In this sector, tasks such as warehouse operations and inventory management—extensions of logistics—may see earlier humanoid adoption. In contrast, tasks including customer-facing and in-store operations during customer hours require higher safety standards, thus humanoid adoption in this field will take some time.

There are also high expectations for the use of humanoids in defense and disaster response, where the goal is to replace humans in dangerous tasks. However, due to the highly unstructured nature of these environments, testing and training are not easy, making practical implementation in these areas particularly challenging. The same characteristics apply to the use of humanoids in space development.

Summary

Rising Attention to Humanoids: Drivers and Outlook

Demand-side pressures from labor shortages, expectations from governments, new business ambitions of tech companies, and advances in AI are all converging to heighten interest in humanoid robots. Because of their human-like form, there is hope that humanoids can be applied to a wide range of tasks traditionally performed by people. However, widespread adoption will require overcoming technical challenges and establishing robust safety standards. It is anticipated that explosive growth in the use of humanoids for logistics and manufacturing may begin in about two years at the earliest.
 

The Situation in Japan: Opportunities and Challenges

In Japan, humanoids are an important area for both public and private sectors, in terms of both their utilization and development. On the utilization side, Japan’s rapidly aging population is certain to result in a shrinking workforce, leading to severe labor shortages in industries such as manufacturing, healthcare, and elder care. Proactively adopting humanoids will be essential for sustaining the nation’s overall GDP.

On the development side, Japan was once a pioneer in humanoid development, but struggled to commercialize its technologies due to budget cuts and economic stagnation. Over the past decade or so, the U.S. and China have surged ahead at remarkable speed. Internationally, there are cases where the public and private sectors are supporting promising domestic startups, and cases like South Korea where the public and private sectors are forming alliances to strengthen competitiveness in the humanoid field.

Against this backdrop, it will be important to watch Japan’s initiatives to develop and utilize humanoids in the coming years.

[Source] (All last accessed: June 12, 2025)

1. NVIDIA "CES 2025: AI Advancing at 'Incredible Pace,' NVIDIA CEO Says"
   https://blogs.nvidia.com/blog/ces-2025-jensen-huang/
2. NVIDIA "What Is a Humanoid Robot?"
   https://www.nvidia.com/en-us/glossary/humanoid-robot/
3. CNBC "Morgan Stanley says humanoid robots will be a $5 trillion market by 2050. How to play it” https://www.cnbc.com/2025/04/29/how-to-play-a-5-trillion-market-for-humanoid-robots-by-2050.html
4. market.us “Global Humanoid Robot Market by Component (Hardware, Software), by Motion Type (Biped, Wheel-drive), by Application (Research & Space Exploration, Education & Entertainment, Personal Assistance & Caregiving, Hospitality, Search & Rescue, Others), Region and Companies – Industry Segment Outlook, Market Assessment, Competition Scenario, Trends and Forecast 2024-2033” https://market.us/report/humanoid-robot-market/
5. GlobalX Official, “The Rise of Humanoid Robots”
   https://globalxetfs.co.jp/research/the-rise-of-humanoids-explained/index.html
6. TECH BRIEFS, “Executive Insights: Humanoid Robots” https://www.techbriefs.com/component/content/article/52810-executive-insights-humanoid-robots
   
7. Reuters, “Elon Musk: 10 billion humanoid robots by 2040 at $20K-$25K each” https://www.reuters.com/technology/elon-musk-10-billion-humanoid-robots-by-2040-20k-25k-each-2024-10-29/
8. Morgan Stanley “The Humanoid 100: Mapping the Humanoid Robot Value Chain” https://advisor.morganstanley.com/john.howard/documents/field/j/jo/john-howard/The_Humanoid_100_-_Mapping_the_Humanoid_Robot_Value_Chain.pdf
9. Official Boston Dynamics "Hyundai Motor Group Completes Acquisition of Boston Dynamics from Softbank"
   https://bostondynamics.com/news/hyundai-motor-group-completes-acquisition-of-boston-dynamics-from-softbank/
10. TechCrunch “Amazon begins testing Agility's Digit robot for warehouse work” https://techcrunch.com/2023/10/18/amazon-begins-testing-agilitys-digit-robot-for-warehouse-work/

Written by

Mina Hammura, Senior Manager, Deloitte Tohmatsu Venture Support Co., Ltd./Deloitte Consulting US San Jose
 

Supervised by

Yuma Saito, Partner,
Deloitte Tohmatsu Venture Support Co., Ltd. (June 16, 2025)

*The above company names, job titles, content, etc. are current as of the publication date.