（編按，抱歉，又來一篇廢文，本篇為測試ChatGTP的文章，不能作為答案來看，甚至連建議都不是。）
問題一：how 37cfr1.130 can be applied to a patent application in prosecution?
只不過~~~是錯的!!!
ChatGPT的答覆是：(1)克服前案核駁；(2)建立專利性（針對說明書欠缺或可實施性）；以及(3)解釋專利範圍。
講得很正經，卻是＂隨意拼湊＂相關法條的結果，都不是37 CFR 1.130的用途！！！（真令人＂反向＂驚訝！）
以下是ChatGPT的答覆原文：
-------------------------------------
37 CFR 1.130 is a provision of the United States Patent and Trademark Office (USPTO) that allows a patent applicant to submit affidavits or declarations to address certain issues that may arise during the examination of a patent application.
Specifically, a patent applicant may submit an affidavit or declaration under 37 CFR 1.130 to:
(1) Overcome a prior art rejection: If the examiner has cited prior art that the applicant believes can be distinguished from the claimed invention, the applicant can submit an affidavit or declaration to provide evidence of the differences between the prior art and the claimed invention.
(2) Establish patentability: If the examiner has raised a rejection based on lack of written description or enablement, the applicant can submit an affidavit or declaration to provide evidence of how the claimed invention meets the requirements of written description or enablement.
(3) Address claim interpretation: If the examiner has taken a different interpretation of a claim term than the applicant intended, the applicant can submit an affidavit or declaration to explain the intended meaning of the claim term.
It's important to note that affidavits or declarations submitted under 37 CFR 1.130 must be supported by evidence such as data, experiments, or expert opinions. They cannot be used to add new matter to the application or to broaden the scope of the claims.
Overall, 37 CFR 1.130 provides a mechanism for patent applicants to provide additional evidence or clarification to the examiner during prosecution in order to overcome rejections and improve the chances of obtaining a patent. Regenerate response
-------------------------------------
繼續問這些答案來源？
問題二：do you know where you got these information?
ChatGPT回覆這個問題來源的回答很經典，這是它從MPEP與CFR"理解"來的，不是copy and paste，但卻不能當作legal advice。
以下是ChatGPT的答覆：
-------------------------------------
As an AI language model, I was trained on a diverse range of textual data, including legal documents such as the United States Code of Federal Regulations (CFR) and the Manual of Patent Examining Procedure (MPEP) published by the United States Patent and Trademark Office (USPTO). The information I provided on 37 CFR 1.130 and its application to patent prosecution is based on my understanding of these legal documents and my training on various other textual sources. However, please note that my responses are purely informational and should not be considered legal advice.
-------------------------------------
my two cents:
這樣看來，ChatGPT或者還不能應付太過嚴肅的事，但將來...？？？（有空來問一下針對特定判例的看法？）
對於"學習"法律、世界大事、維基百科、文學經典...等已經是固定內容的文本，或者是有＂標準答案＂的內容來說，對於「一般電腦程式」來說應該是基本應該要會的，但對於「ChatGPT」等＂自然語言＂聊天AI應該是反而是＂太過嚴肅＂的話題，因此答案經過AI的理解就會有錯誤，AI厲害的是"資料收集並整理與歸納"，但經過本次測試來看，或者聊天機器人是用來對付"複雜與時常變動"的資訊，確實還不能當作法律建議，甚至連資訊正確性都有問題。
ps. 與一般搜尋引擎果然不同，但給出明顯錯誤的答案也是ChatGPT厲害的地方，或者這就是需要擔心的情況！！！
Ron
沒有留言:
張貼留言