Tencent’s internal “AI search recipe” has been made public!

Smart Things Author | Bi Weihao Editor | Mo Ying Enter a very common question in the AI ​​era into the search box - check Tencent's recent stock price, product and member changes. This is Google’s non-AI search results: ▲Google search results In Google’s traditional search, users get a series of web links; in Yuanbao search, the same question presents very different results. ▲The difference between the two search results of WorkBuddy connected to Yuanbao Search just reflects the changes that search is going through. In the past, search engines returned web links, and users had to filter, compare and organize information themselves. In the AI ​​era, search began to directly integrate multiple sources of information and generate answers. The higher requirements for search in the AI ​​era correspond to the two major pain points in the implementation of large models - illusion and knowledge lag. That is, no matter how powerful the model is, it may produce hallucinations or provide outdated information due to untimely knowledge updates. When the Agent that can call tools and perform tasks becomes the focus, errors or outdated information in the retrieval may be passed along with subsequent tool calls, affecting the judgment and execution of the entire task. Therefore, the search must not only be complete and updated in a timely manner, but also be as accurate as possible, and organize the information into results that the Agent can directly use. For this root problem, RAG (Retrieval Augmentation Generation) is one of the important technical paths to alleviate knowledge lag and improve answer reliability. However, if developers and enterprises want to build their own RAG, they need to establish and maintain long-term information source collection, data cleaning, index construction and other aspects. This also means "high cost", and the value of ready-made search API is to lower the access and operation and maintenance threshold. On October 20, Tencent upgraded Yuanbao Search and provided API services to individual developers and enterprises through Tencent Cloud. In the past year, Yuanbao Search has continued to optimize, integrating public resources from the entire network and Tencent's high-quality content, building an index scale of hundreds of billions, and has been tempered by Tencent's internal applications such as WorkBuddy and ima. The significance of this upgrade is not only the release of a search capability, but also that Tencent will