Recently, the State Administration for Market Regulation and the National Development and Reform Commission jointly issued the "Guidelines for Artificial Intelligence Measurement System and Capacity Building (2026 Edition)" (hereinafter referred to as the "Guidelines"), which systematically layout the construction of artificial intelligence measurement capabilities, focus on cracking industry pain points such as algorithm "black boxes" and "data shortages", and strengthen empowerment in 14 key areas such as intelligent manufacturing, smart healthcare, and smart transportation.

Among them, in response to pain points such as algorithm "black box" and poor decision interpretability, the "Guidelines" deploy key technologies such as internal state monitoring and representation of AI systems to promote the establishment of reliable, secure, and trustworthy measurement standards for artificial intelligence, and achieve "measurable, comparable, and traceable" AI technology performance.

Focusing on the problem of "data scarcity", the "Guidelines" clearly propose to build datasets with the highest metrological characteristics, standard reference datasets, and test datasets, establish a basic resource sharing mechanism, break down industry data barriers, achieve data security sharing, and provide accurate "food and forage" for artificial intelligence algorithm training and evaluation.

In addition, the Guidelines also promote the deep integration of metrology technology into 14 key areas such as intelligent manufacturing, smart healthcare, and smart transportation. Research on metrology technology is carried out around key parameters such as the reliability of AI diagnostic algorithms, helping to solve quality assessment problems in industrial digital transformation and effectively enhancing people's sense of security and achievement in the application of artificial intelligence.

The reporter learned that in the next step, relevant departments will build a batch of artificial intelligence measurement technology research and development application centers, which will take the lead in key areas such as smart supervision and smart healthcare, forming replicable and promotable "artificial intelligence+measurement" application scenarios, and accelerating the construction of a measurement support system that matches the development of AI leading industries.