Recently, the first industry standard in the field of embodied intelligence jointly drafted by the China Academy of Information and Communications Technology and more than 40 units has been officially released. This standard has established a unified benchmark testing framework for the field of embodied intelligence, marking a new stage of embodied intelligence evaluation with standards to follow. It is understood that this standard focuses on key foundational technologies of artificial intelligence and benchmark testing methods for embodied intelligence, clarifying the framework and capability requirements for embodied intelligence systems. It will be officially implemented on June 1st. Embodied intelligence is at a critical stage of transitioning from the laboratory to commercial implementation. ”Wei Kai, Director of the Institute of Artificial Intelligence at the China Academy of Information and Communications Technology, said. Data shows that by 2025, there will be over 140 domestic complete machine enterprises and over 330 humanoid robot products released. However, Wei Kai believes that the current development of embodied intelligence industry is facing two major bottlenecks. One is the "workshop style" research and development model. Model tuning and deployment often heavily rely on the experience of algorithm engineers; The data is scattered across various data collection factories, making it difficult to effectively integrate and reuse; Hardware is mostly assembled non-standard and often requires manufacturers to perform "nanny style" maintenance and debugging after delivery, making it difficult to deliver on a large scale. The second is the fragmented ecological pattern. The model performance is strongly bound to the ontology, and changing hardware can easily lead to failure; The system capability highly depends on the practical training effect of specific scenarios, and it is easy to malfunction once the scenario changes; The lack of unified standards in the supply chain leads to difficulty in reusing technological achievements and low efficiency in industrial chain collaboration. The trustworthy embodied intelligence evaluation system is the bridge connecting technology research and large-scale industrial applications, providing a unified technical specification framework for the industry, establishing a foundation of mutual trust, and promoting embodied intelligence from 'workshop style' development to industrial development. ”Wei Kai said. In his view, establishing an evaluation system is to create a trustworthy benchmark for the industry, measuring true intelligence, good products, and strong reliability. On the one hand, by verifying the generalization ability and deployment effectiveness of the model in real scenarios, standards help users distinguish between true intelligence and flashy technology, which can force enterprises to develop truly scalable and replicable embodied basic models. On the other hand, standards enable enterprises to clarify what qualified products are, thereby reducing the cost of technology selection and adaptation, and optimizing the resource allocation of the entire industrial and supply chain. A comprehensive evaluation of the stability and reliability of embodied intelligence in complex environments can also provide security guarantees for large-scale product implementation. ”Wei Kai said, "More importantly, with this trustworthy benchmark, the upstream and downstream of the industrial chain have a common collaborative foundation, and technology research and development, hardware manufacturing, and scenario applications can be effectively connected." The newly released standard specifies the benchmark testing framework, methods, and indicators for embodied intelligence systems in both simulation and real environments. The evaluation system proposed by this standard supports the testing of basic abilities, cognitive reasoning abilities, and full loop closed-loop abilities, covering four methods: static simulation testing, dynamic simulation testing, real environment testing, and combined testing. As the first officially released industry standard in the field of embodied intelligence, the introduction of embodied intelligence benchmark testing methods is of great significance for promoting technological progress, application implementation, and industrial development, and has pointed out the direction for industry development. ”Wei Kai said. At the level of technology research and development, this standard provides a unified measurement basis for the assessment of embodied intelligence capabilities, which can guide the direction of technology iteration, help the industry identify high-value technology routes early, and reduce the ineffective investment of research and development resources; At the application implementation level, provide standard and standardized support for industry users to carry out product selection and application verification, avoiding the phenomenon of "bad money driving out good money"; At the level of industrial development, embodied intelligence will accelerate its transition from the laboratory to real-life scenarios, promoting the engineering and industrial application of embodied intelligence technology achievements. Wei Kai believes that the establishment of independent evaluation standards can help guide the convergence of innovation resources across the country towards the direction that is in line with China's industrial advantages and technological roadmap, avoiding passive catch-up on tracks set by others. In addition, once the evaluation criteria mature and are internationally adopted, it will be beneficial for China's embodied intelligence products, solutions, and platforms to go global, further creating a "Chinese model" in the field of embodied intelligence. (New Society)