In this blog post, as humanoid robots—which resemble humans—are rapidly advancing and entering real-world industrial and care settings, we’ll examine not only the convenience and potential this technology brings but also issues such as safety, cost, employment, and ethics.

From Science Fiction Movies to Real-World Industrial Settings

“Star Wars,” “I, Robot,” “The Terminator,” “Transformers,” “Pacific Rim,” “Iron Man.” You may be wondering why I’m suddenly mentioning the titles of science fiction movies. One thing these works have in common is the appearance of robots that think or move like humans, or machines that resemble humans. In the past, such robots were primarily confined to the realm of science fiction, but today, we have reached a stage where it is not uncommon to encounter humanoid robots in real-world industrial settings. In manufacturing sectors—including the automotive and electronics industries—industrial robots and automated equipment are already in widespread use, and recently, competition has intensified in the development of robots that can perceive their surroundings, make judgments, and move accordingly, thanks to the integration of artificial intelligence and robotics technology. In the automotive industry, in particular, the technology has advanced to the point where humanoid robots are being trialed in actual production processes. BMW tested the use of humanoid robots in its South Carolina plant to move and position parts, and is now evaluating the potential for their use at other production facilities as well. The technology once known as “driverless cars” is also continuing to evolve under the name “autonomous driving technology.” As such, robots are no longer a futuristic technology confined to the movies. If robotics expands into various areas of daily life—such as industry, services, logistics, and caregiving—it will have a significant impact on human lifestyles and work practices. Therefore, it is only natural that social interest in robotics is growing. However, it is still difficult to say with certainty when humanoid robots equipped with general intelligence and physical capabilities comparable to those of humans will actually emerge. While it is clear that the possibilities expand as technology advances, no technology’s advancement necessarily brings only positive outcomes. As technological efficiency and convenience increase, issues related to safety, liability, labor, and ethics may also grow; consequently, discussions surrounding humanoid robots are becoming increasingly important.

The greatest challenge facing humanoid robots is safety

Safety is the primary reason for exercising caution in the development of humanoid robots. In the past, there were insufficient specific safety standards and legal frameworks governing robots; however, today, international standards regulating the safety design and risk mitigation of industrial robots have been established and are continuously being revised. The International Organization for Standardization’s (ISO) safety standards for industrial robots cover not only the safety design of the robots themselves but also the risk mitigation measures required when integrating robot systems into actual work environments. However, the existence of safety standards does not mean that all risks have been eliminated. The U.S. Occupational Safety and Health Administration (OSHA) also explains that robot-related accidents can occur during non-routine tasks such as programming, maintenance, testing, setup, and calibration, and that workers face the risk of injury from unintended movements when they enter a robot’s operational range. In actual industrial settings, serious accidents involving robotic arms or automated equipment are still being reported. Therefore, if humanoid robots are to move freely in the same space as humans and perform more complex tasks, we must consider a wide range of risks—not just simple mechanical failures, but also sensor errors, software defects, communication failures, erroneous AI decisions, and improper user operation. As seen in the past incident at a German automobile factory where a worker was killed by an industrial robot, robot-related accidents do not necessarily result solely from defects in the robot itself. A combination of factors—such as human operational errors, failures in safety zone management, and issues with system design and work procedures—may also play a role. Therefore, it is crucial not to focus solely on science-fiction-style risks, such as the idea that robots might commit crimes on their own, but rather to establish a safety system that realistically accounts for potential malfunctions, erroneous commands, cybersecurity issues, and the ambiguity of liability. In particular, as humanoid robots come to live and work in close proximity to humans, they will inevitably require more sophisticated safety standards and liability frameworks than industrial robots.

Can the Enormous Development Costs Be Justified?

The second issue to consider is cost. As with other cutting-edge technologies, developing humanoid robots and deploying them in actual industrial settings requires significant research and development expenses and capital investment in facilities. In the past, major information technology companies attempted to secure a foothold in the robotics industry by acquiring various robotics firms; today, automotive, artificial intelligence, and robotics companies continue to make large-scale investments in humanoid development. However, it is inappropriate to equate the cost of corporate acquisitions with the total cost of humanoid robot development, as has been suggested in the past, or to conclude that development costs are excessively high based on unverified figures. This is because the economic viability of humanoid robots today cannot be judged solely by the price of a single unit. One must consider not only research and development costs and hardware prices but also the costs associated with AI training, maintenance, batteries and power, safety equipment, software management, and modifications to the work environment. Conversely, if robots perform repetitive tasks for long periods, increase productivity, and reduce the risk of industrial accidents, there is a possibility of recouping the investment costs over the long term. Therefore, rather than unconditionally labeling the costs invested in humanoid robots as a waste, it is necessary to compare them with the benefits that could be achieved by allocating the same resources to other social purposes. Governments should consider the benefits of investing in welfare, safety, and education, while companies should compare the economic viability of introducing humanoid robots with alternatives such as investing in improving working conditions and enhancing the efficiency of existing automation equipment. Ultimately, if humanoids fail to offer clear advantages in efficiency and safety over existing machines or industrial robots, questions will inevitably continue to arise regarding whether investing massive amounts of development funds is economically justified.

Will Robots Take Away Human Jobs?

The third issue is employment. If humanoid robots achieve sufficient safety and economic viability and are commercialized on a large scale, it is highly likely that they will take over some of the tasks previously performed by humans. This is because, from the outset, many robots and automation technologies are designed to have machines take over repetitive or physically demanding labor previously performed by humans. However, automation does not necessarily mean that all human jobs will disappear. Technological advancements have historically reduced existing job roles while creating new ones, and even within the same occupation, they have shifted the balance of tasks handled by humans versus machines. In recent labor market forecasts, while robots and automation are identified as key factors leading to a reduction in certain jobs, it is also projected that new jobs will be created through the combined effects of various factors, including artificial intelligence, automation, the energy transition, and demographic shifts. In particular, robots and autonomous systems not only replace certain jobs but also serve as drivers for the creation of new technical roles as well as maintenance and operations positions. The problem lies in the fact that these changes do not provide equal opportunities for all workers. Production workers, those engaged in simple, repetitive tasks, and workers in fields where automation is relatively easy—such as facility guidance and logistics—may face greater disruption during the job transition process. If humanoid robots become sophisticated enough to replace human hands and feet and adapt to various work environments, the scope of automation is likely to expand even further than it is today. Therefore, when discussing the development of humanoid robots, we must not focus solely on the efficiency of the technology itself but must also consider issues such as retraining, career transition, social safety nets, and the division of labor between humans and robots.

Convenience and productivity are the greatest advantages of humanoid robots.

On the other hand, proponents of humanoid robot development cite the convenience and productivity gains for humans as key advantages. This can be considered one of the most fundamental purposes of robotics technology, including humanoids. In particular, humanoid robots offer different possibilities compared to existing industrial robots, as they can perform tasks previously done by humans without requiring large-scale changes to the spaces and tools humans use. When combined with artificial intelligence and sensor technology, they can recognize their surroundings and perform tasks autonomously within a certain range without requiring humans to manually control every movement. The introduction of automated machinery into manufacturing processes following the Industrial Revolution was also intended to reduce the burden of human labor and increase productivity; today, automated systems are utilized in various industries, including automotive, electronics, logistics, and food. Recently, even humanoid robots have been experimentally deployed in actual automotive production lines to transport and position parts, indicating that the potential for industrial applications of humanoids has reached a much more realistic stage than in the past. However, increased convenience and productivity do not mean that safety issues disappear. While automated systems can reduce human error, they can also create new forms of system failures and problems related to human-machine interaction. Therefore, describing past aircraft accidents simply as “accidents caused by the malfunction of automated equipment” may be an oversimplification. Aircraft accidents often result from a complex interplay of multiple factors, including pilot judgment, training, the design of automated systems, the condition of sensors and equipment, and operational environments. The same applies to humanoid robots. Rather than viewing automation itself as dangerous, it is far more important to consider how humans and automated systems interact, how to transition to a safe state when errors occur, and who should hold ultimate responsibility and control.

Opening New Possibilities in Dangerous Labor and Care Sectors

Another rationale for developing humanoid robots is their ability to take over tasks that are dangerous or difficult for humans to perform. If robots can be deployed not only for simple, repetitive tasks that humans tend to avoid but also in environments hazardous to humans—such as those involving fire, disasters, radiation, extreme heat, and toxic substances—they could significantly help reduce loss of life. In fact, industrial robots are already widely used to perform dangerous, repetitive, and unpleasant tasks. In the past, it was difficult and extremely costly to develop robots capable of performing complex tasks in hazardous environments; however, with advances in artificial intelligence, computer vision, force control, sensors, and battery technology, the range of tasks robots can perform continues to expand. Nevertheless, in disaster sites where unpredictable situations occur in rapid succession or in complex industrial environments, it remains challenging to reliably operate robots equipped with human-level judgment and agility. Therefore, alongside deploying humanoid robots for hazardous tasks, efforts must also be made to improve the safety of existing work environments and strengthen accident prevention systems. This is because deploying robots does not automatically eliminate risks to humans; rather, the nature of the risks and the methods for managing them may change.
The potential applications of humanoid and social robots are also being researched in fields where human interaction is crucial, such as elderly care, rehabilitation, education, and psychological support. PARO, Japan’s seal-shaped therapeutic robot—a well-known example for many years—has been used to support emotional interaction among the elderly and patients with dementia. Recent studies also report that such care robots may have a positive impact on anxiety, restlessness, social interaction, and the burden of caregiving. However, the results across studies are not consistent, and further verification is needed regarding the magnitude and durability of these effects. In particular, it is more realistic to view care robots as tools that assist human caregivers rather than as complete replacements for them. In situations where caregivers cannot always remain by a patient’s side, robots can help reduce the burden of care by performing tasks such as monitoring, conversing, providing simple guidance, and assisting with routine duties. However, given that those receiving care are physically and emotionally vulnerable, the potential harm resulting from malfunctions, privacy breaches, or erroneous judgments must be scrutinized even more rigorously. Furthermore, ethical issues—such as the extent to which it is desirable for robots to replace human emotional relationships and whether excessive reliance on robots might weaken human relationships—must also be considered.

Why Safety Must Take Precedence Over Efficiency

As such, the debate surrounding the development of humanoid robots involves both arguments for and against. However, it is not necessary to view this issue solely through the two extreme lenses of “development” versus “halting development.” The key lies in determining the acceptable level of risk while maximizing the benefits of humanoid robots, as well as deciding what level of cost and efficiency to accept in order to reduce that risk. Pursuing maximum efficiency means expanding the technology’s performance and scope of application while accepting a certain level of risk, whereas prioritizing safety means ensuring the technology is thoroughly verified and equipped with safety measures before use, even if this results in a somewhat slower adoption rate or lower efficiency. Today’s robot safety standards are also evolving away from the premise that risk can be completely eliminated, toward analyzing risks from the design stage and reducing them to the greatest extent possible. This principle is particularly important for humanoid robots, which move in the same space as humans and are capable of direct physical contact. This is because a single technical flaw can result not only in economic loss but also in harm to human life and physical well-being.
Therefore, rather than taking an approach that insists on the unconditional development of humanoid robots or, conversely, calls for a complete halt to all development, it is important to make safety and accountability prerequisites for technological advancement. In the past, the risks associated with humanoid robots were primarily linked to science fiction scenarios—such as rebellions or artificial intelligence escaping human control—but the problems we actually face today are far more concrete. These include sensor and software errors, collisions between workers and robots, cybersecurity, privacy protection, liability in the event of accidents, changes in the labor market, and issues of humanity in caregiving. At the same time, humanoid robots have the potential to take over dangerous work that is difficult for humans to perform, reduce the burden of repetitive tasks, and assist with caregiving challenges in an aging society. Ultimately, rather than labeling humanoid technology itself as good or evil, what matters is the purpose and criteria for its development, as well as the extent to which it will remain under human control. If the harm that could result from the technology’s shortcomings directly threatens human life and safety, we must prioritize thorough verification and safety measures—even if it means sacrificing some efficiency—rather than taking risks in pursuit of maximum efficiency. For humanoid robots to remain a technology that serves humanity, what is needed is not unconditional rapid development, but rather an approach that ensures safety standards, legal liability, ethical principles, and measures to address labor transitions evolve in tandem with the pace of technological advancement. In this sense, humanoid robots remain a double-edged sword even today, and which edge is directed toward human society depends not so much on the technology itself as on the choices made by the humans who develop and use it.