The State Administration for Market Regulation recently released three national guidance documents on AI in the steel industry — the "AI Steel-Industry Application Scenario Classification Guide," the "AI Steel Large-Model Dataset Technical Requirements," and the "AI Steel Large-Model Service-Capability Maturity Assessment." Together they form a closed loop of "scenario classification – dataset norms – maturity assessment," aimed at using large-model technology to solve complex steel-industry problems and support enterprises' digital-intelligence transformation, moving steel AI from single-point exploration to systematic, measurable deployment.
The three standards mark a new stage where steel-industry AI application shifts from "single-point exploration" to "systematic and measurable." They clarify deployment boundaries through scenario classification, cement the model-quality foundation through dataset norms, and provide selection and acceptance yardsticks through maturity assessment — expected to lower enterprises' repeated trial-and-error and integration costs.
On October 19, nine departments including the Ministry of Commerce issued opinions on accelerating "technology-prospering commerce." The document proposes strengthening R&D and conversion, accelerating key-technology R&D in areas such as new displays and smart terminals, and expanding typical application scenarios and viable business models. It calls for cultivating commercial technology-application innovation centers, focusing on consumer goods to build technology-transfer institutions, and accelerating market conversion of frontier technologies such as brain-computer interfaces and embodied intelligence, while supporting smart warehousing and digital supply chains.
The policy signal — stronger than a single subsidy — aims to unblock the R&D–scenario–business-model chain, with technology-transfer institutions and innovation centers likely becoming new intermediaries.
Source: State Administration for Market Regulation / Ministry of Commerce
