Kagi's approach to AI in search
16 Mar, 2023
We’d like to discuss how we see AI’s role in search, what are the challenges and our AI integration philosophy. Finally, we will be going over the features we are launching today.
During those early days, we were passionate about AI’s potential, particularly in question-answering and summarization. This was when the Transformer model, which paved the way for modern LLMs (large language models), was just invented in 2017. However, the initial models were small and pale compared to the immense LLMs today.
In those early days, it took a lot of creativity and innovation to extract the full potential of AI. Today, the power of LLMs such as GPT is so great that “the hottest new programming language is English,” as Andrej Karpathy put it. This has democratized the ability to “create magic,” allowing even small companies to compete with industry giants, as we demonstrated later in this text.
Following 2019, we placed most of these endeavors on hold to concentrate on developing the finest search product. Our early efforts, however, enabled us to use machine learning to deliver answers to questions directly within the search experience, making us the only search engine besides Google and Bing able to do so when we launched the public beta in 2022..
