RAG(Retrieval-Augmented Generation;検索拡張生成)は、外部のデータベースから必要な情報を検索・取得し、LLMが事前学習していない情報も回答できるようにする手法で、社内にある技術情報などを参照することにより、生成AIの回答精度を高めることができるため、特に技術開発においては他社との競争力の源泉となることが期待されています。 しかし、RAG導入が思い通りに進んでいない会社が多いようなので、生成AIに調査させました。 Gemini(Google), ChatGPT(OpenAI), Perplexity, GensparkのDeep Research機能による調査結果を添付しましたので、ご参照ください。 なお、生成AIによる調査・分析結果は、公開された情報からだけの分析であり、必ずしも実情を示したものではないこと、誤った情報も含まれていることについてはご留意されたうえで、ご参照ください。 RAG Implementation Not Progressing as Expected Retrieval-Augmented Generation (RAG) is a technique that enables Large Language Models (LLMs) to search and retrieve necessary information from external databases, allowing them to answer questions even about topics they were not pre-trained on. By referencing internal technical data, RAG can significantly improve the accuracy of generative AI outputs and is expected to become a key source of competitive advantage, especially in the field of technology development. However, many companies appear to be struggling to implement RAG as intended. To investigate this issue, I asked generative AI systems to conduct research. Please refer to the attached results from the Deep Research functions of Gemini (Google), ChatGPT (OpenAI), Perplexity, and Genspark. Please note that the research and analysis conducted by generative AI are based solely on publicly available information and may not fully reflect the actual situation. The results may also contain inaccuracies, so please use them with caution. Your browser does not support viewing this document. Click here to download the document. Your browser does not support viewing this document. Click here to download the document. Your browser does not support viewing this document. Click here to download the document. Your browser does not support viewing this document. Click here to download the document.
0 Comments
Leave a Reply. |
著者萬秀憲 アーカイブ
July 2025
カテゴリー |