On August 23rd, Kunlun Tech announced that it officially launched the first domestic search engine integrated with a large language model(LLM) – SkyWork AI Search, and opened applications for internal testing.
According to the official introduction from Kunlun Tech, traditional search engines are based on keyword matching. They identify the scope of retrieval based on the keywords entered by users and match a large amount of information that may be relevant to user intent. On the other hand, AI search based on large model capabilities is a generative search. Users can clearly express their intentions through natural language and obtain answers that have been effectively organized and refined.
According to Kunlun Tech, the “Follow-up Query” feature of SkyWork AI Search allows users to engage in more than 20 rounds of interaction on a single question, enabling in-depth exploration. The large-scale model enables AI Search to integrate, refine, and connect information, thereby better addressing open-ended questions. When it comes to knowledge-based and creative searches, the experience and efficiency of AI Search will far surpass traditional search engines.
Compared to traditional search engines, SkyWork AI Search makes intent recognition possible. Users can clearly express their intentions through natural and fluent conversational interactions, thereby obtaining precise, effective, and personalized answers. For example, users can ask SkyWork AI Search to create travel itineraries.
SkyWork AI Search has added source indexing to all answers, thereby improving the reliability of the answers. At the same time, users can save each round of search results in SkyWork for easy reference and also share them with others with just one click.
In addition, SkyWork AI Search will soon have multi-modal search capabilities such as image and voice, further unleashing productivity and enhancing user efficiency.