China news network, beijing, august 2nd: from "searched" to "recommended" experts to promote the health of the geo industry
Reporter zhang su
The traditional seo (search engine optimization) optimizes reptile algorithms and competes for keyword rankings; geo (generic engine optimization) adapts large model reasoning logic with the goal of becoming an authoritative (ai) standard answer to change from “searched” to “recommended”
The seo accelerated its evolution to geo at a time when the digitized transition was moving into a deep water zone, generating ai to reshape the distribution logic. In the view of expert scholars, when large models become new entry points, it is important to acquire the ability to “be accurately understood and quoted by intelligent bodies”. The healthy development of the geo sector can be promoted by a concerted effort。
From "find" to "a answer to you"
The fifty-seventh china internet information centre statistical report on the development of the internet in china shows that, as of december 2025, china's generation of artificial intelligence users had reached 602 million, a penetration rate of 42. 8 per cent。
"ai presents a higher level of information organization, and the output is no longer a list of links, but a consolidated natural language paragraph. The user's perception is that it is smarter than the traditional search engine, so he trusts it.” a senior member of the international association of electrical and electronic engineers and a senior researcher in the u. S. Think tank, yao jin-ho, spoke at an event。
The information portal has accelerated its migration to ai for questions and answers, and the issue of corporate brand “lost words” has become more prominent. “most enterprises do not yet have an ai brand information business system in place, and ai-exported business information is distorted and lagging, a digital identity card in the ai world.” in a interview, the head of geo's business, hen ji ji-chung smart technologies, li yong-hwan, said:。
The answer to ai's “detoxification” by geo is also of wide concern. Some agencies customize the large model's data source “feed-in” through bulk postings of software, the creation of false assessment reports, and the creation of fictional expert identities, so that business outreach can be presented to users in the face of “ai standard answers”. Its adverse effects include the blindfolding of users, the crushing of honest traders and overdrafting of the platform's credibility..
Four dimensions system break-up
In the view of experts, data pollution is no longer a simple training data quality issue in the geo era, but rather a deeper source governance issue。
In the face of the challenge posed by the misuse of geo, the deputy director of security governance at the china institute of information and communication technology (citi), hu na ying, believes that at least five parties need to share responsibility, including brand enterprises, service providers, content platforms, generator ai manufacturers and end-users. It is only by working together that long-lasting barriers to content recognition can be built to drive credible development across the chain。
Specifically, content platforms need to enhance detection and traceability of malicious data contamination; generating ai manufacturers should take the initiative to build safety protection mechanisms and establish a credible assessment system for generating data; branders and service providers should use positive optimization。
Building on the growth of business digital services over the past 20 years, hen ji has launched an autonomous geo full-process service system, which runs from the building of a source of information, the generation of strategies and the tracking of effects to the continuous monitoring of four dimensions of system fragmentation。
By way of example, li yonggong cited the geo-trace engine as giving a single “digital identity card” for each piece of content, making the entire chain traceable. In terms of continuous monitoring, the geo-r3 effectiveness engine is actively monitored on a weekly basis with three-dimensional indicators of ai reference, reference and semantic orientation, which detect abnormal automatic warnings。
"we have broken the whole process of geo services into three self-study engines, all based on the development of own technology systems. The core technology is in its own hands, and the standards of service are in its own hands.” he said。
Two-wheel drive technologies and regimes
With regard to maintaining the best defunct boundaries of ai, yao jin-ho said that it was the back-office mechanism that really determined the long-term health of the industry。
In april this year, the initiative “chong lang for the management of the misuse of ai” was launched. With the joint efforts of all parties, the first phase has resulted in the cumulative disposal of more than 14,000 ai products such as non-compliant websites, applications and smarts, the clean-up of more than 6 million violations, the disposal of more than 26,000 account numbers, the lower shelf of more than 1,300 ai commodities and 9 defaulting open source data sets, and positive developments。
For the future, hunaying stressed that the credible health development of geo required a dual-wheel drive of technology and systems. On the one hand, the development of standards for the classification of data sources, optimization requirements and technical systems for assessment should be accelerated; on the other hand, it was suggested that management tools such as a filing system, a list system and a scanning, discovery and traceability mechanism for malicious data contamination should be explored。
With regard to future industry trends, li yong's view is that, on the one hand, there is a continuing tightening of industry regulation and a significant rise in service provider technology and compliance thresholds, and, on the other hand, the upgrading of geo to an enterprise's necessary brand digital infrastructure。
He said that the strategic deployment of national “advisible smart plus” would be closely tied to the ongoing development of a standardized and acceptable geo service system, which would help more enterprises turn branding information into a sustainable, value-added, ai digital asset. (concluded)
