At ICRA 2026, the IEEE Robotics and Automation Society panel discusses humanoid robotics’ return on investment, reliability, safety, deployment and investment, with Andrea Kay of Silicon Valley Robotics, systems engineer Katina, and other panelists and audience members.
Andrea Kay describes Silicon Valley Robotics as a nonprofit industry association founded in 2010 to support robotics innovation and commercialization. She says robotics investment has risen alongside venture capital and has since recovered at a faster rate, making robotics a venture-investable category. She also points to funding choices as a factor in which technologies reach broad social impact.
Katina says ROI depends on the customer’s use case, profitability and the costs faced by both manufacturers and integrators. Her company plans to begin with five to ten robots doing focused tasks, then scale toward 10,000 at customer sites in Europe within four years. An audience member from Rockwell cautions that 200 hours of operation is modest for a fleet, while another robotics engineer recommends detailed logs, failure analysis and continuous joint testing to find recurring hardware problems.
Panelists distinguish near-term workplace value, such as robots completing part of a person’s work, from investment based on future capability and falling component costs. One panelist says cheap humanoids can be little more than remotely operated hardware, since capable software remains costly and difficult. Katina adds that manufacturing, service, installation and safety costs matter alongside the bill of materials.
Katina says her company is starting with a wheeled humanoid platform because its safety case is simpler than a dynamically stable biped’s, especially around people in industrial settings. The panel discusses possible benefits of legs, including maneuverability and compact use of space, while acknowledging that emergency-stop behavior and standards remain unresolved. Andrea Kay argues that different applications may call for different robot configurations, including wheeled platforms with arms and a head-like attention indicator.
Panelists debate ambitious forecasts, with estimates ranging from about 40,000 humanoids in 2026 to one million by 2030, while noting that production capacity and customer demand remain open questions. Katina says her team is prioritizing specific industrial tasks and plans to extend capabilities based on customer needs; other panelists expect autonomy, teleoperation and easier teaching methods to coexist. In home settings, panelists emphasize that safety, reliable performance and a clear use case will affect what people are willing to pay.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Diana Wolf Torres, robotics newsletter writer at Droids and chief evangelist at Lightwheel, discusses physical AI and robotics at NVIDIA GTC 2026 with host Jonathan Stevens on the Thinking Machine podcast.
Artium Solos, founder and CEO of Humanoid, discusses the company’s plan to build and deploy industrial AI humanoid robots with its teams in Vancouver, Boston, and London.
Humanoid’s founder and robotics team discuss building the HMND 01 wheeled humanoid, from its commercial purpose and design to integrating and preparing the prototype for customer trials across Boston, London, and Vancouver.
