Zhuxin, a subsidiary of Changying Data, builds brand trust infrastructure for the AI era.

When the search entry shifts from the traditional "link list" to AI agents that directly output answers, the phenomenon of enterprises' brands being "silenced" or "misquoted" in AI responses has become a common pain point for B2B enterprises, and new market demands are spawning a brand-new track. Based on the judgment that "trust comes from consensus, and recognition from business partners is the most core trust endorsement", Changying Data launched the CastTrust project at the beginning of this year, completed company registration in Chenzhou, Hunan in July this year, and officially entered the GEO (Generative Engine Optimization) service track. With the Mutual Trust Index, TruthPage, Mutual Trust Chain Verification Network and MCP Trust Protocol as its core products, it helps enterprises build verifiable, quotable and sustainable brand trust assets in the AI agent ecosystem.

 AI becomes the new search entry, while enterprises are "silenced" in the answers generated by agents 

At present, AI search is in a stage of rapid penetration, and the industry pain points are becoming increasingly prominent: on the one hand, the vast majority of Chinese small and medium-sized enterprises have no content layout targeting AI agents, so when users consult relevant questions, their brands cannot appear in AI-generated answers, missing out on potential business opportunities; on the other hand, the industry is generally faced with the problem of "distribution deviation between API and C-end answers". Most service providers only focus on optimizing API indicators, ignoring the SystemPrompt, RAG private knowledge base and post-processing links added when AI agents generate C-end answers, which is equivalent to practicing on the wrong target, resulting in a huge deviation between the final optimization effect and the real user experience. Different from the industry's GEO logic that simply chases traffic, the core value of CastTrust is to help enterprises synchronously cover their authoritative information to the publicly trusted sources that AI can crawl and the real C-end answer distribution, filling the market gap of enterprise brand trust infrastructure in the AI era. 

With the Mutual Trust Index and mutual verification network, "trust" is encapsulated as a callable interface 

At the initial stage of the project, the founding team targeted a differentiated development path, and its core technical highlights are concentrated on the exclusive product of the Mutual Trust Index, for which there are no similar players on the market at present. Different from the logic of other GEO service providers that only optimize traffic, the Mutual Trust Index created by CastTrust is China's first five-level scoring system that quantifies "the degree of an enterprise being trusted by AI", which carries out materialized scoring from 14 thematic dimensions, with results that are comparable, traceable, interpretable and auditable. On this basis, the project has built a complete product closed loop of "diagnosis → truth page → mutual trust chain → continuous monitoring", built a self-owned C-end calibration layer for the distribution deviation problem to simulate the real answer distribution of C-end agents, with semantic similarity reaching 80-90%, making the monitoring results closer to what users actually see; at the same time, it adopts the MCP/Skill plugin architecture to encapsulate "trust" as a standard interface that can be called by agents, truly realizing "trust as API"; the Mutual Trust Chain Verification Network relies on dofollow mutual link endorsements from upstream and downstream business partners and BingBacklinks verification scoring, replacing self-promotion with real mutual proof relationships to build EEAT-style public authoritative footprints, and the enterprise mutual verification data can flow credibly within the authorized scope to support rating output. 

Overlaid with regional agents and chamber of commerce association flywheel, the early commercialization path has been verified 

At present, the project is still in its embryonic stage. The core GEO system and Mutual Trust Index system have been built, and it is about to enter the market promotion stage. The core team has a total of 5 members, who have mature advantages in technical R&D, business logic design, chamber of commerce resource connection, market promotion and other links. In the process of project advancement, the team made it clear that cold start is the core challenge at the current stage, and will rely on its accumulated national chamber of commerce resources to open up the market. At present, 5 chamber of commerce associations have reached preliminary cooperation intentions, and the signing is expected to be completed this month. 

According to current calculations, there are about 40 million small and medium-sized enterprises in China that need to build awareness through AI search. Combined with government and enterprise, chamber of commerce customers that can be reached through the regional agent network, the GEO track is still in the early stage, and the overall market is a 100-billion-level untapped market. As the penetration rate of AI search continues to rise, market demand will continue to increase. Targeting this market space, Changying Data has created a clear layered business model. In the early stage, it will obtain revenue through GEO services and enterprise digital intelligent services, with layered pricing starting from 9,800 yuan/year. At the same time, it will develop city and industry regional agents to output standardized services, and rely on chamber of commerce resources to form a customer acquisition flywheel of "free chain entry → GEO referral commission → government and enterprise resource connection"; in the later stage, it will charge service fees for agents' calls and queries of the Mutual Trust Index to complete the business closed loop.

In the future, Changying Data will continue to polish the Mutual Trust Index engine, gradually expand coverage to more industry themes, open the standard version of the MCP Trust Protocol to the public, rapidly expand the scale of member sites on the Mutual Trust Chain, set up multiple regional agent outlets, and steadily expand the paid customer base. The long-term goal is to become an indispensable trust infrastructure service provider in the AI agent ecosystem, reduce transaction and communication costs between enterprises in the agent era, and promote the compliant and orderly flow of trusted data. Some heads of chamber of commerce associations who have contacted the project in advance said that small and medium-sized enterprises generally lack the ability to layout brand trust in the AI era, and CastTrust's products accurately hit the demands of the vast number of member enterprises, which has high practical value for helping enterprises connect to AI traffic and expand business opportunities. 

Talking about the entrepreneurial journey of the project, the founding team of Changying Data said that the essence of trust in the AI era is credible data. Changying Data insists on building down-to-earth, lightweight trust infrastructure, refuses over-design and high-cost stacking, and hopes to help Chinese enterprises seize the new opportunities of AI search in the most economical way, gain a firm foothold in the brand-new traffic environment, and establish their own long-term trust assets.