Annual humanoid robot shipments are expected to exceed 700,000 units by 2035, generating revenues of approximately $15 billion, according to a new report from Interact Analysis. The research firm projects that near-term shipments will remain below 100,000 units annually, with the sector unlikely to reach broad commercial viability before 2032. Current demand is being sustained primarily through pilot projects, government subsidies, and strategic industry partnerships rather than self-sustaining commercial adoption.
The forecast presents a more conservative timeline than some industry projections but aligns with the structural constraints that practitioners and researchers have consistently identified as limiting deployment scale.
What Is Holding the Market Back
Interact Analysis identified technology readiness as the primary constraint. The report cites gaps in embodied AI capability, severe data scarcity, insufficient hardware durability, and inconsistent manufacturing quality as the technical barriers that must be addressed before industrial-scale deployment becomes economically viable. On the ecosystem side, safety standards, certification pathways, and insurance mechanisms remain underdeveloped – frameworks that enterprise customers need before committing to large-scale humanoid procurement.
“The market is shifting from hype to pragmatism, with vendors and early adopters prioritising operational stability over headline specifications,” said Marco Wang, research analyst at Interact Analysis. “Wheeled platforms are emerging as the preferred near-term form factor for real-world industrial deployment” – a direct acknowledgment that the bipedal humanoid, while commercially significant in the long run, is not necessarily the most practical configuration for immediate industrial use.
Where Growth Will Come From
Industrial manufacturing and warehousing are identified as the primary near-term deployment environments, given their structured operating conditions and concentration of early adopters willing to manage integration complexity. Public service applications in China are also emerging as an early category, driven by government-supported initiatives rather than commercial demand.
Household applications are expected to remain limited in the short term due to safety concerns and the operational complexity of unpredictable home environments – consistent with the assessment from GigaAI and Unitree’s own leadership that domestic deployment remains challenging even as companies target it for the 2027-2028 period.
The China-U.S. Dynamic
Interact Analysis expects China to account for more than 65% of real-world application shipments by 2035, driven by government investment programs, subsidies, and procurement by state-owned enterprises. The United States is expected to expand through continued AI investment, capital availability, and the cost pressure of comparatively high domestic labor costs. Together, the two countries are forecast to represent more than 85% of global humanoid robot demand by 2035.
The robot training and data collection segment is projected to grow near-term before stabilizing as simulation technologies improve, reducing the need for physical robot data gathering. Real-world commercial applications are forecast to grow from approximately 10% of total production in 2025 to the dominant segment by 2035 – a trajectory that implies the current period of subsidized pilots and strategic partnerships is laying the foundation for a commercial market that has not yet arrived at meaningful scale.
