• Home
  • Services
  • About
  • Contact
  • Blog
  • 知財活動のROICへの貢献
  • 生成AIを活用した知財戦略の策定方法
  • 生成AIとの「壁打ち」で、新たな発明を創出する方法

​
​よろず知財コンサルティングのブログ

リリース後のPerplexity Patentsの評判

10/11/2025

0 Comments

 
2025年10月30日にPerplexity Patentsがリリースされてからしばらくたちました。評判はどうでしょうか?生成AIに調べさせました。
各種生成AI(Grok4/Felo AI/Claude/Perplexity/ChatGPT/Gemini)が公開レポートで収集・要約した声をみると、全体としては「革新的で使いやすいが、実務の最終判断には慎重に」という評価に収れんしており、特に初期評判は概ねポジティブだが、ベータ段階ゆえ精度検証が鍵、というトーンが共通でした。
肯定的な反応は、「自然言語・会話型・セマンティック検索により、非専門家でも当たりを付けられる入口が広がった。」「初期の先行技術調査、競合・トレンド把握が短時間で進むとの評価。」「特許に加えて論文・OSS・ブログ等のNPLも横断でき、探索の網が広がる。」「回答に原典リンクが並ぶ点は信頼性面の長所と評価。」「操作性・検索範囲・引用優先を長所として整理。」
否定的/慎重な反応は、「FTOや無効資料・訴訟用途は不適、専門家検証の併用が必須(「予備調査向け」止まり)。」「被引用回数など詳細データの誤り指摘、検索ログや監査証跡の不足、データカバレッジの不透明さ、日本語精度の改善余地。」「開発元の著作権訴訟・スクレイピング批判等への目配りが必要との声。」「結果の偏り・網羅性への不安、無料版のクォータ/速度面などの実務上の制約。」
ひとことで言うと、「入口(予備調査・学習・方向付け)には強いが、出口(法的結論)には弱い。使うなら“AIで広く当てる→専門家と公式DBで詰める”の二段構えが前提」というのが各レポートの共通見解のようです。
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。
 
Perplexity Patentsがリリース
https://www.perplexity.ai/ja/hub/blog/introducing-perplexity-patents
 
Perplexity Patents: 誰でも使える AI 搭載特許検索
 
 
Reputation of Perplexity Patents After Its Release
It has been some time since Perplexity Patents was released on October 30, 2025. How has it been received? I asked several generative AI models to investigate.
Based on reports compiled and summarized by multiple generative AIs (Grok 4, Felo AI, Claude, Perplexity, ChatGPT, Gemini), overall sentiment has converged around the view that “it is innovative and user-friendly, but professional judgment is still required for final decisions.” In general, early impressions are mostly positive, but since the product is still in beta, accuracy verification is considered key.
Positive reactions include:
  • “Natural-language, conversational, and semantic search expands access even for non-experts.”
  • “Highly rated for speeding up preliminary prior-art searches and competitive or trend analyses.”
  • “Covers not only patents but also papers, OSS, and blogs, broadening the scope of discovery.”
  • “Providing source links alongside answers improves reliability.”
  • “Praised for usability, search coverage, and prioritization of citations.”
Critical or cautious reactions include:
  • “Not suitable for FTO, invalidation, or litigation use; professional verification remains essential (best limited to preliminary research).”
  • “Some point out errors in citation counts and detailed data, lack of search logs or audit trails, unclear data coverage, and room for improvement in Japanese accuracy.”
  • “Calls for attention to ongoing copyright lawsuits and criticism of data-scraping practices against the developer.”
  • “Concerns about bias, comprehensiveness, and practical constraints such as quotas and speed limits in the free version.”
In short: “Strong at the entry stage (preliminary research, learning, direction-setting), but weak at the exit stage (legal conclusions). Effective use requires a two-step approach: ‘cast a wide net with AI → verify with experts and official databases.’”
Please note that the findings above are based solely on publicly available information gathered by generative AIs and may not fully reflect the actual situation; some inaccuracies may be included.

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.
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

    September 2025
    August 2025
    July 2025
    June 2025
    May 2025
    April 2025
    March 2025
    February 2025
    January 2025
    December 2024
    November 2024
    October 2024
    September 2024
    August 2024
    July 2024
    June 2024
    May 2024
    April 2024
    March 2024
    February 2024
    January 2024
    December 2023
    November 2023
    October 2023
    September 2023
    August 2023
    July 2023
    June 2023
    May 2023
    April 2023
    March 2023
    February 2023
    January 2023
    December 2022
    November 2022
    October 2022
    September 2022
    August 2022
    July 2022
    June 2022
    May 2022
    April 2022
    March 2022
    February 2022
    January 2022
    December 2021
    November 2021
    October 2021
    September 2021
    August 2021
    July 2021
    June 2021
    May 2021
    April 2021
    March 2021
    February 2021
    January 2021
    December 2020
    November 2020
    October 2020
    September 2020
    August 2020
    July 2020
    June 2020

    カテゴリー

    All

    RSS Feed

Copyright © よろず知財戦略コンサルティング All Rights Reserved.
サイトはWeeblyにより提供され、お名前.comにより管理されています
  • Home
  • Services
  • About
  • Contact
  • Blog
  • 知財活動のROICへの貢献
  • 生成AIを活用した知財戦略の策定方法
  • 生成AIとの「壁打ち」で、新たな発明を創出する方法