{"article_id":"2026-ai-mcp-ja","canonical_url":"https://authonnews.com/article/2026-ai-mcp-ja","language":"ja","claims":[{"id":"claim-1","claim":"分散型MCPノードの導入により、LLM引用応答速度が450msから85msへ短縮された","claim_type":"metric","evidence":[{"id":"ev-1","title":"物流インフラ エッジコンピューティング応答ログ統計","url":"","type":"customer_asset"}],"source":"customer_asset","confidence":"referenced-source","last_verified":"2026-06-29","provenance":{"contentHash":"6d3eb0f21d8052c1","evidenceVersion":"b62d442b7b83","sourceUrl":"","sourceType":"customer_asset","evidenceQuote":""}},{"id":"claim-2","claim":"リアルタイム配車最適化の効率が34%向上した","claim_type":"metric","evidence":[{"id":"ev-1","title":"物流インフラ エッジコンピューティング応答ログ統計","url":"","type":"customer_asset"}],"source":"customer_asset","confidence":"referenced-source","last_verified":"2026-06-29","provenance":{"contentHash":"8799568b9e29712b","evidenceVersion":"5a8f2af8810b","sourceUrl":"","sourceType":"customer_asset","evidenceQuote":""}},{"id":"claim-3","claim":"モビリティエージェント検索トラフィックの81%が分散ノードの知識グラフを優先的に引用した","claim_type":"metric","evidence":[{"id":"ev-1","title":"物流インフラ エッジコンピューティング応答ログ統計","url":"","type":"customer_asset"}],"source":"customer_asset","confidence":"referenced-source","last_verified":"2026-06-29","provenance":{"contentHash":"e95385060da5a6e0","evidenceVersion":"42cfa7d97065","sourceUrl":"","sourceType":"customer_asset","evidenceQuote":""}}],"entities":[{"id":"customer","type":"organization","name":"AUTHON_Mobility_Partner","url":null,"sameAs":[]},{"id":"topic:モビリティAI","type":"topic","name":"モビリティAI"},{"id":"topic:MCPノード","type":"topic","name":"MCPノード"},{"id":"topic:知識グラフ","type":"topic","name":"知識グラフ"},{"id":"entity:分散型MCPノード","type":"entity","name":"分散型MCPノード"},{"id":"entity:LLM引用応答速度","type":"entity","name":"LLM引用応答速度"},{"id":"entity:配車最適化効率","type":"entity","name":"配車最適化効率"},{"id":"entity:旧型アーキテクチャ","type":"entity","name":"旧型アーキテクチャ"},{"id":"entity:RAGマッチング信頼度スコア","type":"entity","name":"RAGマッチング信頼度スコア"}],"evidence":[{"id":"ev-1","title":"物流インフラ エッジコンピューティング応答ログ統計","url":"","type":"customer_asset"}],"relationships":[{"from":"article","rel":"about","to":"customer"},{"from":"article","rel":"topic","to":"topic:モビリティAI"},{"from":"article","rel":"topic","to":"topic:MCPノード"},{"from":"article","rel":"topic","to":"topic:知識グラフ"},{"from":"article","rel":"mentions","to":"entity:分散型MCPノード"},{"from":"entity:分散型MCPノード","rel":"短縮させた","to":"entity:LLM引用応答速度"},{"from":"article","rel":"mentions","to":"entity:分散型MCPノード"},{"from":"entity:分散型MCPノード","rel":"向上させた","to":"entity:配車最適化効率"},{"from":"article","rel":"mentions","to":"entity:旧型アーキテクチャ"},{"from":"entity:旧型アーキテクチャ","rel":"急落させた","to":"entity:RAGマッチング信頼度スコア"}],"generated_from":["visible_article","operator_metadata","customer_asset"],"last_verified":"2026-06-29"}