Press Relase / News Release Distribution Service [@Press]

Note: This page is a machine translation of the Japanese original and is provided for reference only.
In the event of any discrepancy between this page and the original, the original shall prevail. Click here for the original text.

Lawyers Check Legal Mistakes by ChatGPT and Other AIs for Free Launched the beta version of "AI Legal Judge" for corporations and sole proprietors.

○/X/Unable to judge by attorneys. New service to mitigate the risk of self-judgment of generated AI.

リーガルブレスD法律事務所

Legal Breath D Law Office (Location: Chuo-ku, Osaka; Representative Attorney: Shinichi Yuhara) has launched a new service, "AI Legal Judge (free screening beta version)," which provides free judgments by lawyers on the correctness or incorrectness of legal consultation answers given to generated AI such as ChatGPT. The new service, "AI Legal Judge (free screening beta version)," will be available on a trial basis for corporations and sole proprietors from June 2025.


Using the generated AI, the service asks legal questions and consultations such as, "Please draw up an outsourcing agreement," and "Does the Freelance Law not apply to xx transactions?" and other legal questions and consultations, and the number of businesspersons who believe and act on the answers as they are is rapidly increasing. In reality, however, there are many answers that seem plausible but are legally incorrect (AI error), and this is becoming a risk in practice.


In order to prevent legal accidents caused by such AI errors, this service allows users to submit legal advice given to the AI and the content of the answers, and lawyers will make a simple judgment by selecting "○ (correct)," "× (erroneous)," or "impossible to judge.



Background and Social Issues

While the use of generative AI for legal advice and contract checking assistance will expand after 2023, risks due to reliance on AI answers are becoming apparent.

For example, AI responses may contain "legally incorrect" content, such as the following

In some cases, AI's answers may contain "legally incorrect" content, such as the following: - Contracts created by AI violate laws and regulations.

The AI always answers that the agreement between the parties is always valid.

A situation in which the AI gives advice on how to deal with a situation based on a court case that does not exist.

In an increasing number of cases, business operations are conducted without being aware of such wrong answers, resulting in problems later on.



Outline of the service

Service name: AI Legal Judge (free screening beta version)

Target: Corporations and sole proprietors who consulted the generated AI for legal advice (*Consumers are not eligible for this service)

Description : In response to AI's answers,

       Content :A lawyer makes a simple judgment of correctness or incorrectness by answering "○/×/unable to judge" to the AI's response.

Target AI: Major generated AI such as ChatGPT, Claude, Google Gemini, etc.

Fee :Free of charge (up to 1 case per day)

Judgment format: Judgment results only (no reason explanations or correction suggestions are included)

How to use : Send questions and AI answers using the dedicated form.

       (Judgment will be returned within 24 hours from 10:00 a.m. on the next business day.)


For more information about the service and free application, please visit the following website.

https://media.ys-law.jp/information/aijudgment-screening/



代表弁護士 湯原 伸一

Representative Attorney Shinichi Yuhara


AI法務ジャッジ

AI Legal Judge


Firm Profile

Firm name: Legal Breath D Law Office

Representative : Shinichi Yuhara

Location : Honmachi TD Bldg. 10F, 4-2-12 Kyutaromachi, Chuo-ku, Osaka 541-0056, Japan

Establishment : September 2012

Business : Lawyer

URL : https://www.ys-law.jp/

Image

Logo Image