In mid-October, a wave of global AI-agent data and domestic policy signaled that agents are moving from concept prototypes to deliverable, billable enterprise "digital labor." On the policy side, the "15th Five-Year Plan" outline lists agents and embodied intelligence as core AI innovation directions, while the "Implementation Opinions on Regulating and Developing Agent Applications" — China's first national-level agent-specific document — sets a framework, and the State Council's "AI+" campaign targets over 70% agent penetration by 2027.
Deployment is broadening across sectors. The financial industry is being called the "year one of financial agents," with more than ten brokerages connecting third-party AI platforms; Industrial Securities launched an AI wealth-management app enabling full natural-language operation from signal to trade execution, and Zheshang Securities began compliance-security testing of its agents. On the enterprise side, Anta introduced agents to transform global e-commerce, cutting manual entry by about 90%, while神州控股科捷 launched a supply-chain agent that digitizes veteran employees' know-how.
Marketingforce (02556.HK), a Hong Kong-listed agent company, reported that in the first three quarters its AI-application revenue grew about 114% year on year, sustaining triple-digit growth on a high base, with precision-marketing services up 32%. Its "full-stack Token factory" strategy lowers the barrier and cost for enterprises to build agents.
As the industry moves from pilot to scale, the ability to commercialize enterprise AI is beginning to diverge among vendors.
Source: China Daily
