Digital Extraordinary Administrative Advisory Committee Study Team (8th) of the Legal Affairs Office of the digitalization Working Group

Overview

Date and Time: Thursday, May 12, 2023 (2023) from 10:45 to 12:00
Location: Online
Agenda:
1. Opening
2. Proceedings
  1. Efforts to digitize review of digital legislation and official gazettes
  2. Survey and demonstration of the development and utilization of digitalization and law data for legal affairs, AI utilization, etc.
  3. Questions and Answers and Exchange of
3. Adjournment

Materials

Minutes

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister Now that it is on time, we would like to hold the 8th meeting of the "digitalization Review Team for Legal Affairs".

My name is Mr. Yamaguchi, and I will be chairing the meeting today. I am Counsellor Digital Agency. Nice to meet you.

Members and observers are invited to participate online today. Thank you for taking the time to attend.

The previous meeting was held in November last year, so there is a little time. As you can see on the screen, there are two main topics on today's agenda. Since the meeting was held last year, in review of digital legislation, for example, relevant bills have been submitted to the Diet, and discussions are under way to legislate the initiative to digitize official gazettes.

In addition, with regard to the legal work of digitalization and others, we are ready to start demonstration this fiscal year. Today, I would like to report on the progress of our efforts since the meeting last year, explain our future policies for each Issue, and I would like to hear your frank opinions.

I would like to introduce the people who will participate as observers today. Until now, people from related ministries and agencies have participated as observers, but from today, the Cabinet Legislation Bureau and the Cabinet Office will participate as observers. Thank you very much.

If you don't mind, first of all, Mr. Kitamura, Inspector of the General Affairs Office of the Director-General of the Cabinet Legislation Bureau, I would appreciate it if you could give me a few words of greeting. Thank you very much.

Cabinet Legislation Bureau: This is Kitamura of the Cabinet Legislation Bureau. Thank you for your kind invitation. I will be participating as an observer from this time.

We, the Cabinet Legislation Bureau, have been established in the Cabinet as an organization to assist the Cabinet from a legislative perspective. We are in charge of examining bills, cabinet orders, and treaties drafted by each ministry and agency.
With regard to legal affairs in digitalization, the Cabinet Legislation Bureau will work in cooperation with related ministries and agencies, including Digital Agency. Thank you very much.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .
Next, Mr. Yoshida, Director of the official gazettes Digitization Study Office, General Affairs Division, Minister's Secretariat, Cabinet Office, I would appreciate it if you could give me a few words. Thank you.

Official gazettes Digitization Study Office: My name is Yoshida, and I am the Director of the official gazettes Digitization Study Office in the General Affairs Division of the Secretariat of the Cabinet Office in . Nice to meet you.

We have been discussing the computerization of official gazettes since last year, and under the policy of submitting it to the extraordinary Diet session this year, we launched an expert panel in March this year and are considering it. We will hold the first three sessions, the fourth session will be held within this month, and the report will be compiled in the middle of the year, and we will make efforts toward the extraordinary Diet session.

To this end, we have established a study office in the General Affairs Division since April 1 this year, and are working to establish an appropriate system. We will proceed with this in consultation with Digital Agency, the Printing Bureau, and other related organizations. Thank you very much.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Now, I would like to move on to the first part of the agenda. The secretariat will explain our efforts to digitize review of digital legislation and official gazettes. Mr. Nakano, Director of Planning, nice to meet you.

Secretariat (Nakano): Thank you very much Digital Rincho Secretariat.

Now, I would like to explain about agenda 1.

I would like to review the materials. First of all, I would like to project two pages on review of digital legislation. Since the last meeting in November, various efforts have been made in review of digital legislation, and I would like to explain the background to date.

With regard to review of digital legislation, you discussed the "Guidelines for Confirmation of Compliance with the Digital Principles" at this meeting in September last year. We have started trial implementation of the guidelines for the bills scheduled to be submitted to the extraordinary Diet last year, and have since expanded the content of the guidelines and are conducting review of digital legislation for the bills scheduled to be submitted to the ordinary Diet.

I would like to briefly reflect on this. In the small part below, in August last year, we implemented this initiative on a trial basis for the bills to be submitted to the extraordinary session of the Diet in the fall of 2022, bringing forward the review of digital legislation Plan, which was to be implemented on a trial basis among the bills to be submitted to the ordinary session of the Diet in 2024.

In addition, in December last year, with regard to provision, which could be related to regulations on paper and in-person processes, the results of the inspection of the bills to be submitted to the extraordinary Diet were announced, along with the timing of the development of notifications and circulars clarifying that digital technologies can be used. Since December last year, based on the inspection of the bills to be submitted to the extraordinary Diet, the content has been expanded, and review of digital legislation is being implemented for the bills to be submitted to the ordinary Diet.

The specific content of the expansion is the yellow shaded part in the figure on the right, but there are places where it is difficult to understand the actual situation of regulations on paper and in-person processes just by looking at the text of the bill. When applying the phases, including the lower law and notifications, etc., the process will be clarified, that is, the process of when the notifications and notifications and lower law will be developed will be clarified, digitalization will be implemented using Technology Map and the Technology Catalog, and Digital Agency will implement the necessary support for validation, such as cross-sectional digital technology, in the future. We are currently implementing review of digital legislation for the bill to be submitted to the ordinary Diet session.

With regard to the future of review of digital legislation, the outline of the bill is shown on the next page and the next page, and the The Act to Advance Reforms of Digital Regulations bill was submitted to the National Assembly in March this year. It includes the amendment of the Basic Act on the Formation of digital society, which stipulates digital regulatory reform as a national basic policy, and the Digital Legislation Bureau process, which includes provisions related to the review of digital legislation process.

The regulations stipulate that necessary measures must be taken to ensure that digital society does not hinder the effective use of digital technologies based on the progress of digital technologies as basic policies for the formulation of measures for the formation of regulation, and that this review will be positioned as a matter to be described in Priority plan.

Going forward, after the enactment of this bill, I would like to continuously and autonomously make efforts in review of digital legislation as an initiative positioned in the law.

To be specific, the Cabinet Office and each Ministry will make appropriate use of the law and Technology Catalog when planning a new Technology Map, etc., conduct inspections based on the Guidelines for Confirmation of Compliance with the Digital Principles so that a regulations on paper and in-person processes will not be newly specified, and submit the results of the inspections to Digital Agency. In addition, Digital Agency will implement the review of digital legislation while maintaining the necessary systems.

Pages 4 and 5 are the outline of the bill.

Next, I would like to explain official gazettes's efforts for digitization. This is what the people of the official gazettes Digitization Study Office of the Cabinet Office, who have been participating as observers since today, are working on. First of all, regarding page 7, this is a partial update of the materials submitted to the Digital Rincho last December, and we are working to realize an electronic official gazettes. It has been a symbolic initiative in digitalization, Japan, to digitize official gazettes, which has been issued in paper form since the Meiji Era, and we believe that the digitization of official gazettes, which is also a means of promulgating the law, will be the foundation of DX in the legal field.

As for the background, as we have covered in this study team so far, the original copy of the official gazettes is conventionally used as a paper medium, and it is impossible to abolish written documents in administrative procedures or reuse data. Therefore, as Reform 1, we are making it unnecessary to submit a paper official gazettes in administrative procedures, and as Reform 2, we are positioning an electronic official gazettes as the original copy of the official gazettes by the new law on the issuance of the official gazettes in the first place.

The outline of the specific process is as shown in the figure at the lower left. By the way, in foreign countries, the EU, France, and Germany are making efforts to make the electronic version of official gazettes the original.

With regard to the specific status of implementation of Reforms (I) and (ii), on Page 9, regarding Reform (I), on January 27 this year, the Cabinet approved the use of the Internet version of official gazettes in administrative procedures. The Commercial Registration Act and other laws have so far only allowed the submission of paper documents, and it was necessary to actually purchase and submit a official gazettes. However, the Cabinet approved this, and the same identity between the paper official gazettes and the Internet version of official gazettes has been secured, and the Government has already implemented the reform to allow the submission of the Internet version of official gazettes.

As for the next slide, in addition to that, regarding Reform (2), in order to position the electronic official gazettes as the original version, the General Affairs Division of the Minister's Secretariat of the Cabinet Office, which is participating as an observer today, has established the official gazettes Reform Review Team, and has started discussions on a new law concerning the issuance of the official gazettes. At that time, a meeting called the official gazettes Electronic Review Meeting has been held and discussions have been advanced.

As stated in the second box below, there is no law concerning official gazettes, and it is based on common law. Therefore, it is necessary to amend the common law to issue official gazettes by electromagnetic means, and to enact a new law concerning the issuance of law by the official gazettes, and we are conducting consideration on this matter. official gazettes

The next slide shows the materials submitted to the official gazettes Electronic Review Meeting in April. As shown in the figure on the left, the current paper official gazettes is considered to be issued by official gazettes when it is posted on the bulletin board of the main office of the Printing Bureau in Toranomon. It will be delivered to official gazettes sales offices nationwide and can be viewed, and although it is not positioned as an official official gazettes, it will be possible to view the Internet version of official gazettes at the same time. The current mechanism of official gazettes is being considered to be changed as shown in the figure on the right. The final approval of official gazettes will be uploaded to a dedicated server and can be viewed anywhere for free through the Internet. When it becomes accessible on the Internet, it will be issued by official gazettes.

However, as a response to people who cannot use the Internet, official gazettes will be made available for viewing on digital signage, etc. at the official gazettes posting location at the same time as it is posted on the Internet. However, this does not affect the official gazettes issuance time, and we are considering a mechanism in which official gazettes Record Matters will be delivered and sold to those who wish to receive them.

With regard to the next slide, in particular, I would like you to discuss the utilization of electronic official gazettes today, and with regard to the future plan that we are considering, as decided in the schedule, we will complete the examination and discussion points by the middle of this year on the mechanism to digitize official gazettes, which has been issued in paper form mainly by the Cabinet Office, and submit the bill to the Diet as soon as possible. You just mentioned the extraordinary Diet session, but it will be submitted to the Diet.

In considering the bill, we will implement a system design that will enable the following operations. As the Prime Minister stated in the Digital Policy Consultation, we will keep in mind that we will abolish paper official gazettes in the future. We will not simply replace the current paper official gazettes with an electronic official gazettes, but will build a technology-neutral mechanism that can respond to future technological innovation.
In addition, as you have been discussing in this study team, we will provide machine-readable electronic official gazettes data in order to utilize official gazettes data. It is important to have a mechanism that can store official gazettes for a long period of time, and we will provide authentic information because it has not been tampered with and law will be promulgated. Also, as you will discuss later, we believe that we should proceed with BPR of administrative work related to official gazettes in cooperation with e-LAWS. We would appreciate if you could discuss this point.

That's all from me.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Then, I would like to set aside about 10 minutes for a question and answer session and an exchange of opinions. If you have any questions or opinions, please let me know by clicking the button of a show of hands. Thank you very much.

Thank you, Mr. Yoneda.

Yoneda Member: Kagoshima University.

As a citizen of Japan, I would like to express my deepest gratitude for the vigorous efforts you are making to computerize official gazettes and the examination of the new and amended laws.

Based on the history so far, I would like to express my opinion to the extent that I am thinking. I would like to comment on three points that I would like to emphasize.

Regarding official gazettes, the legislation of digitalization is being advanced, but as long as digitalization is assumed, it is very important to take into account future technological development, and I believe that law's regulations must not create vendor lock-in or become a hindrance to responding to technological progress. In that sense, I would like to ask you to advance to ensure technological neutrality.

In addition, with regard to the utilization of official gazettes information, in fact, how it is used is not sufficiently understood, and I believe that it is not well known. However, official gazettes contains not only information on law, but also a variety of information, including personnel information, government procurement information such as bidding cases and successful bids, and information on corporate financial statements and statutory public notices. I am aware that these are very actively used.

In particular, as researchers, we hope to be able to use electronic data from official gazettes for historical and social scientific research. As I mentioned earlier, government procurement information and corporate financial results announcements are used as value-added information in private sector.

Among them, there are some that are used in a certain inappropriate manner, such as information on bankrupt persons, but they should be dealt with in the direction of post-control based on the results of utilization, or in the direction of imposing a regulation or considering the utilization of judicial systems such as civil lawsuits. At present, I believe that it is necessary to protect the inherent roles of the past and encourage development in the aspect of utilization so that information will not be reduced as much as it has been in the past, just as information will be provided in official gazettes as usual.

Finally, I would like to comment on the necessity of long-term implementation. As for law, of course, it serves to publicly communicate and record the behavior of the country itself, and I believe that it is necessary to satisfy three requirements: it must be stored as a record document for a long period of time, it must be stored in a state that can be viewed, and it must be in a form that can be utilized. This may be a slight misexpression, but I believe that both the information on official gazettes thus far and the information on official gazettes in the future must be kept in a "state of long-term preservation by society."

At present, I feel that the emphasis of digitalization's discussions on legislation is on digitalization and communicating it on the spot, but I believe that it is necessary to consider official gazettes as something that will remain for future generations while taking into consideration future technological development.
That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister . Thank you very much for all the important points.

Mr. Nakano, what can I do for you?

Secretariat (Nakano): Thank you very much , thank you very much.

You pointed out three points regarding official gazettes digitization. First of all, regarding technology neutrality, at the official gazettes Digitization Review Conference held by the Cabinet Office, it was stated that the specific technology related to the issuance of official gazettes and the data format of official gazettes will be legally made technology neutral so that it can respond to future technological innovation. In this direction, based on Mr. Yoneda's comments, the Cabinet Office would like to consider it, and if necessary, Digital Agency would like to cooperate.

Second, regarding the utilization of official gazettes information, as I have stated on page 8 of the materials I have submitted today, it includes not only information on law, but also, as you pointed out, personnel changes and company financial results announcements, and I believe that it is utilized in the form of information reorganization by researchers and people in private sector, so Digital Agency would like to cooperate with the people in charge in the Cabinet Office so that the reuse of such information will not be hindered.

As you pointed out, long-term storage is a theme that has not been taken up by the Study Team. I am aware that it will be taken up by the Cabinet Office's official gazettes Digitization Study Meeting in the future, but we would like to exchange opinions with related organizations, consider what kind of state is desirable, and proceed with consideration together with the people of the Cabinet Office.

Thank you very much for your valuable comments.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

We have received a show of hands from the members, but we would like to designate them one by one.

First, Mr. Tsunoda, please.

Member: Mr. Yoneda has just organized the points of discussion, so I will speak in that order.

The first point was the issue of technology neutrality, I think. Of course, I also talked about various things, such as preventing vendor lock-in, but in addition, data tends to be interpreted in a variety of ways, so even in such a case, it is desirable to specify the data part according to various interpretations, to make sense of it, and to realize it in a form that can be removed from the outside later, at an appropriate granularity, and without unnecessary control information. Of course, it is good to proceed in the form of XML tagging, but in terms of the division method used there, I think it is necessary to have a design that can be used for various purposes. By doing so, I think we can say that technology neutrality is maintained.

As for the second issue, the need to utilize official gazettes information, I would like to say this from the perspective of scholars around me. First of all, for example, I myself have had the opportunity to stick to official gazettes for about 10 years, and I have actually used it as a database for how to write revised articles.

In addition, in the case of historical research such as legal history, there are issues that need to be considered, such as the transfer status of public figures and personal data, and in such cases, it is grateful that the data presented in official gazettes can be used with peace of mind. It is possible to trace the public status of public figures and information on decorations, so I think it will be important for scholars who will analyze history in the future. In addition, I think it is also effective that contractors who are newly bidding for the government can understand the status of orders from the government in the most official form. At least, I think the needs for utilization are from the perspective of scholars.

Finally, regarding the issue of the necessity of long-term storage, I would like to emphasize this point the most. What I would like to say in particular is that I specialized in security in a research project I belonged to in recent years, and I have been working with teachers who have been studying security for decades. At that time, what the teachers discussed was the problem of security and the compromise of cryptography. All of the current electronic signature technologies that maintain authenticity are made of cryptography, but there is a problem that the cryptography is compromised and becomes ineffective. What is compromised is that as technology advances, the cryptography itself deteriorates and becomes meaningless in the first place. For example, there has been a caesar cipher method that shifts one alphabet character since ancient times, but it is too simple to be called cryptography and cannot be adopted as Electronic signature technology at present. Even the current robust cryptography method becomes obsolete like this. It is too late to deal with the original data after it becomes obsolete, so I think it is necessary to think carefully.

In particular, research on quantum computers is being conducted in development, and it is said that it will be dangerous if they are truly put into practical use. In that case, it may be okay if it is five to ten years, but when we consider ultra-long-term storage of 20, 30, 40, or 100 years, it will be compromised, that is, electronic signatures will become meaningless. In consideration of such a case, I think it is necessary not only to use a fixed signature encryption method, but also to review it periodically once every few decades, for example, in the form of Shikinen Sengu of Ise Jingu Shrine, if it is a system that has been in existence for more than 1,000 years, or to make it possible to review the encryption technology and electronic signature technology to be used each time a possibility of compromise is seen. In the first place, it is a new technology that has not been seen yet, so it is a problem that cannot be covered by technology, so I would like to add one provision to cover it properly in terms of legal development and to make it possible to transition more and more new ones.

Of course, as for past data, we should retroactively convert it into an electronic signature because we are afraid of falsification. If there is a possibility of compromise as I said, I think it is necessary to quickly overwrite it with a new encryption method.

As I said earlier, when we are talking about the necessity of long-term storage, as a scholar, when we want to go back to the past and examine various kinds of data, we can say that official gazettes is the first and most reliable evidence. So, I would like to see a robust institutional mechanism that can ensure ultra-long-term storage. That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Then, I would like to ask you to speak one by one. If you don't mind, Mr. Fujiwara, please.

FUJIWARA Member: I'm FUJIWARA. Thank you very much.

I would like to make two comments related to official gazettes. The first is about the format. Although there is technology neutrality, which you have already talked about, the content itself has been based on vertical documents, including the content of the revised law. I feel that if I am not careful, I will end up just posting vertical PDFs. I should probably stop doing that, but I think that if the actual state of the product itself is not changed, there is a possibility that the XML will not be as you imagine. I think I should be careful about that.

Another point is that there have already been discussions in various places about personal data, and in short, the advantages and disadvantages of disclosure. I think there have been various stories about the famous Bankrupt Map. This is not only about official gazettes, but also about the address of the representative of the registration of the corporation of the company, which has been a topic of conversation recently. In the end, it is necessary to record this information for specificity, and it must be reachable in an emergency, but it is common in the sense that it may not be necessary to keep this information out. Therefore, I think it would be good to solve this problem together when official gazettes is digitized. I think the mechanism should be the same, so I don't have a solution to the point where I can say what I should do, but I think it is okay to think about it.

That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Next, Mr. Watanabe, may I have your attention?

Watanabe, Member: Thank you, Thank you for your support. I have three points.

First, I would like to talk about the history from November last year to the present, second, I would like to talk about the evaluation of utilization related to this case, and third, I would like to talk about what the legal department looks like as a practitioner.

First of all, I would like to thank the members of the Cabinet Legislation Bureau and the Cabinet Office for their participation as observers. The period has been vacant since November, but during this period, the Secretariat has worked very hard on the compilation of the bill.

Second, the utilization of electronic official gazettes this time is quite consistent with what we have discussed at the Review Meeting so far, and I would like to highly evaluate it. In particular, I believe that it should be highly evaluated that the reform has actually taken shape in conjunction with the Digital Policy Consultation and other movements.

In addition, I happened to discuss this very story with Professor Shishido of the University of Tokyo at the Agile Governance Symposium of Kyoto University in April. I recognize that the electronic official gazettes is currently undergoing a very complex work of two stages, first determining customary laws and then legislating them. Therefore, I would like to tell you that I have great respect for your efforts and consideration, even though this point is four letters in the form of digitization.

Finally, I would like to make a third point. From the perspective of the Legal Department, as Mr. Fujiwara mentioned earlier, it has been difficult to access official gazettes. However, I would be very grateful if this digitization would make it easier to access Issue in a timely and real-time manner. Exactly according to Mr. Yoneda's research, efficiency, the current No. 1 Japanese legal department, is the official gazettes of legal affairs and IT, and so many companies are struggling. I also hope that this electronic Tokyo will be used as one of many databases.

As mentioned above, I strongly hope that you will advance this reform under various leadership, so I would like to ask for your continued support.

That's all from me.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister Thank you very much for your very encouraging comments.
Next, Mr. Yasuno, please.

Yasuno Member: Just one point for . Thank you for your cooperation.

What I would like to say in this talk is that from the perspective of business operators and engineers, official gazettes will be digitalization and considerable consideration will be given to machine readability, so I recognize that this is a very grateful story.

I would like to ask a simple question. On the upper right of page 11, there is a place where official gazettes can be viewed for free anywhere on the Internet, and a second place where it can be downloaded for a certain period of time. I would like to ask if there is a distinction between viewing and downloading. In addition, as Mr. Yoneda mentioned earlier, it is important to be able to view for a long period of time, so I would like to ask how you think about the period of time, if you have any assumptions.

Regarding the second suggestion, as Dr. Tsunoda said, I think that the compromise of electronic signatures in the encryption part in long-term storage is one Issue, and in the field of electronic contracts, electronic signatures are sequentially applied in the form of long-term signatures, so this is a useful reference. In addition, I think that the update system in the case of critical vulnerabilities in emergency encryption algorithms should be discussed in advance, so I would like you to consider it.

That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Regarding your question on page 11, Mr. Nakano, I would like to ask for your comments.

Secretariat (Nakano): Thank you very much Thank you for your various comments. It is taking a little time, so I would like to apologize for the inconvenience. Regarding page 11, which can be viewed for free and can be downloaded for a certain period of time, when official gazettes is issued and law and others are promulgated, it is of course essential to be able to view them at the timing of their promulgation. As is the case with the current Internet version of official gazettes, it is possible to view things from the past for a certain period of time, and law is planned based on this information. There are issues such as how much information that is stored in the past can be viewed, for example, what to do with this certain period of time. In addition, I believe that it is a discussion that should be worked out in the future about how the electronic official gazettes will be operated by the issuing organization of the electronic electronic bulletin board system, and what will be done because system updates will be necessary in some cases. Thank you for your extremely valuable comments.

Yasuno Member: Just one point for I understand very well. Thank you very much.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Next, Mr. Yagita, please.

Yagita Member: Nice to meet you. I'm Yagita from Legalscape. Thank you very much.

As Mr. Yasuno said earlier, I think it is grateful that official gazettes is computerized from the perspective of a business operator or an engineer. Thank you very much.

As you pointed out various major issues, I would like to comment on one specific matter. Regarding the second part from the bottom on page 12, which is the provision of authentic information that has not been falsified, I believe there are two main topics.

One is whether or not the server providing the Internet version of official gazettes provides the correct information, and the second is whether or not the file downloaded from the server providing the Internet version of official gazettes provides the correct information. I feel that these two stories are mixed now.

In short, the first problem is the security of the government website, which cracks the national server and sends false information. The second problem is that the information was correct when it was downloaded, but the PDF file is changed later. I think there are two problems. If we could build a system in which people who need so-called authentic information can download it directly from the national server and obtain it directly, I think there would be no need to particularly consider the problem of PDF files being tampered with in the first place.

The problem that Dr. Tsunoda mentioned earlier is that the PDF file in (2) is digitally signed, and the strength of the encryption of the digital signature, or security, will decrease in 20 years, so I think it will be tampered with.

However, as long as we can guarantee that the national server will always transmit correct information, whether today, 20 years from now, or 50 years from now, we do not need to store it for a long period of time. I am sorry, but I may be a bit of an amateur, but I believe that you have already been considering this. I would like to ask what your thoughts are on this point.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

May I have a brief comment from Mr. Nakano, Director of Planning?

Secretariat (Nakano): Thank you very much .

Basically, I believe that the technologies used in the Internet version of official gazettes are electronic signatures and time stamps, and for example, if they are tampered with when accessed, when the people view them, green becomes a red lamp, and they can be recognized as tampered with. However, what kind of mechanism the electronic official gazettes will be after the law is actually promulgated and enforced is currently being discussed at this Review Conference. I am a little uneducated about the point you made that long-term storage is not necessary if the server can be guaranteed to be correct even after downloading the file, and I think it is a new point of view, so I would like to discuss it with the Cabinet Office and the National Printing Bureau. I am very sorry that I do not have an answer here.

Yagita Member: I understand. Thank you.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Mr. Yoneda, Mr. Tsunoda, I think you have raised your hand again, but if you don't mind, please start with Mr. Yoneda. Due to the time, I would appreciate it very much if you could summarize your remarks in a compact form. Thank you very much.

Yoneda Member: I would like to make a brief statement.

First of all, the issue of preservation is not limited to data alone. It has been stored on paper in other organization such as the National Diet Library and the National Archives of Japan, but even if it becomes data, I would like you to consider sharing the responsibility for preservation, consider whether it is really okay to store only data, and consider promoting cooperation with related organization.

The other is to respond to the development of technology. Even if we do not go to the level of encryption, it is actually a problem in the Web Archiving Project (WARP Project) of the National Diet Library that we cannot read what has been stored until now because the specifications of the browser have changed recently. I would like to comment that it is necessary to consider that aspect.

That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Then, Mr. Tsunoda, please.

Member: There seems to have been a slight misunderstanding, so in terms of Yagita's story, what I said was not a story after distribution in PDFs and the like, but a story about what will be done if the source of the source is not kept unchanged and what the government is thinking is correct and is publishing is wrong. Since it is the original data that is the source, I would like to emphasize that point.

Of course, Member Yagita also divided it accurately and gave his opinion as the first half, but it is exactly the first half.

That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Thank you very much for your wide-ranging opinions and comments. Based on the comments you made today, we will continue to advance the details of the study.
I would appreciate it if you could move on to the next agenda item because time is short. As for the second item on the agenda, I would like you to discuss the survey and demonstration of the development and utilization of digitalization and law data for legal affairs, and the utilization of AI.

First, Assistant Yamauchi, who is in charge, will explain.

Secretariat (Yamauchi): Thank you for your visit to Secretariat. Based on Material 2, I would like to explain "Survey and demonstration on the development and utilization of digitalization and law data for legal affairs, AI utilization, etc.".

From page 2 onwards, I will continue to look back for a while, but at this meeting, I would like to explain the progress since the previous meeting and our future policies. There are two main points of progress. The first is that the procurement of the survey and demonstration projects has been completed and we are finally at the stage of starting the validation. The second is that AI products using large-scale language models have made rapid progress in the past few months, and it has become necessary to review their suitability for legal affairs. Today, I will explain these two points.

I would like to move to page 7. First, I would like to talk about the survey and demonstration of the development and utilization of digitalization and law data for legal affairs. Please see page 8. Based on the requirements discussed by the previous meeting and the direction of utilization of law data, since the previous meeting, we have summarized the implementation items on the left, examined the specifications, and proceeded with the procurement procedures. The procurement was completed around the end of last month, and as you can see on the lower right, we will conduct the survey and demonstration projects with Daiichi Hoki Co., Ltd. as the managing company. I would like to explain each of the implementation items on the left. There are three groups.

The first is the survey of legal affairs, law, etc. We will survey and analyze the current status, examples, and needs of legal affairs, law, etc., and design and simulate workflows and data structures. Through this, as you can see on the right, we will establish the basis for design and evaluation of legal affairs systems.

The second group on the left is the investigation, examination, and prototyping of the legal affairs system. Here, I have mentioned three points. Regarding the overall architecture, the editing and checking functions of law and other data, and the disclosure and utilization functions of law and other data, as you can see on page 9, they are displayed in different colors. I hope you will understand that the overall architecture corresponds to the overall architecture, the left side corresponds to the editing and checking functions of law and other data, and the right side corresponds to the disclosure and utilization functions of law and other data.

I would like to return to page 8. Going back to the left side, regarding the editing and checking function of law and other data in the middle, I would like to proceed with the assumption of user law at the site of actual testing planning.

My third question is that in the disclosure and utilization of law and other data, we expect interactive design and trial use with users of APIs and the like. For example, we would like to consider actually performing service development using an API prototype and feeding back the findings to design. As a result, as you can see on the right, our purpose is to establish the feasibility and basic design of the system for POC and system development that are expected to be actually introduced next fiscal year or later, to foster understanding of the legal and administrative system through examination with users of the system and users of law and other data, and to establish an environment that will enable constructive design and smooth transition in the future.

Finally, the third group on the left is surveys and research on the current status and future of digital legislation. We are conducting surveys and research on the future vision of advanced technology utilization and the elaboration of a digital legislation roadmap. Here, we would like to conduct surveys and research with researchers and students at universities and so on. As a result, as shown on the right, we would like to clarify the long-term effects of legal affairs digitalization, establish the significance and long-term goals of system upgrading, establish policies for realizing effective policy formulation by utilizing law and other data in the future, and foster interest in research development and human resource development.

So far, we have explained the three groups of implementation items.

On the next page, as you saw earlier, we have organized the overall image of the system for your reference.

The next page, page 10, is an image of the schedule for each implementation item. As you discussed at the meeting before last, there are various examination Issue in the examination of the legal affairs system. Therefore, we would like to take a method of repeating the cycle of prototyping and evaluation by dividing the entire system into several cycles, rather than making it once.

Pages 11 to 15 are reference materials, but detailed items are extracted from the actual specifications. I will not give you an oral explanation.

The above is the outline of the investigation and demonstration to be started.

Now, I would like to move on to page 16. Next, I would like to talk about the second issue that has been discussed since the last meeting, which is the short-term validation of legal affairs assistance using AI and other technologies.

Let's go to page 17. It is the first page on the left. In the last meeting on November 9, in the following few months, AI products using the large-scale language model LLM, and products with dramatically improved performances, were announced one after another, and have been widely talked about. Among them, ChatGPT has attracted attention to the extent that it has been reported that the number of monthly users has exceeded 100 million in the two months since the announcement.

The second point is that in terms of performance, for example, GPT-4 is reported to have scored in the top 10% in a simulation of the U.S. Bar Examination. The third point is that in terms of usability, ChatGPT can be operated only by natural language input (prompt) without writing code, and the output can be obtained as natural language. It is characterized by operability that can be used just like chat.

I will move to the right. In response to the progress of these products, there is also a response in the industrial and public sectors. First, the Liberal Democratic Party's Project Team on AI Evolution and implementation compiled a white paper in April, in which it proposed to immediately start multiple pilot projects that can show results in a short period of time, and one of the specific examples is the provision of legal aid. It has been discussed in other meetings, and just yesterday, AI Strategy Council was newly held.

In addition, the second from the bottom, there are reports that central government agencies and local governments are considering using it, and last but not least, it has been confirmed that legal tech companies are also announcing new services using ChatGPT and others.

Based on these trends, considering the relationship between legal affairs and digitalization, the purpose of legal affairs efficiency is to utilize digital technology and digitalization business, so I believe it is significant to consider the suitability of AI using LLM for legal affairs.

Let's move to page 18. Regarding AI using LLM, it is known that while it is convenient, there are some points to be noted. The first point on the left is a language model. It is originally a model trained to generate natural sentences and exchanges. It is attracting attention for its flexibility and adaptability to various tasks, but calculation and logical inference are not certain by itself, and search is not performed. However, methods to improve it by devising prompts and combining it with external APIs are also being studied.

Based on these characteristics, considerations from the perspective of legal affairs are written on the right. While linguistic tasks such as summarization are expected to be effective, tasks such as retrieval and analysis of law and articles require the law database API, which is the basis. Based on this, processing combining APIs is considered.

It is the second point on the left. In principle, whether the output results are true or appropriate is another matter, so when using the output, we believe that the user needs sufficient knowledge and technology to analyze and judge the accuracy and validity of the content.

Moving to the right, it is considered that sufficient skills such as these are necessary from the perspective of legal affairs, but based on that premise, it is expected to be effective for breast applications that output ideas that are difficult to notice.

This is the third point on the left. I also think that it is necessary to be more careful not to be confused by falsetto. I think that it is necessary to be particularly careful to use objective and critical analysis, but I will move to the right. In particular, in the task of legal affairs, strictness in complex situations is particularly required, so it is considered that a mechanism for appropriately calling attention is necessary. In addition, in order to strictly apply the rules, it may be effective to use another deterministic check function in addition to the language model.

Let's move to page 19. Based on the above, we would like to make a short-term validation for legal affairs assistance using AI, etc. using LLM, and I would like to explain it as a draft. There is a purpose and outline on the left, but the purpose is to sort out the suitability of AI, etc. using LLM for legal affairs, what is likely to be realized immediately, and what needs to be considered in the medium to long term. In this way, we would like to consider the ideal way of utilization.

To be specific, as mentioned in the second part, we will consider using existing products to demonstrate the use ideas that are likely to be realized immediately. As a result, we would like to collect the problems and Issue of the results of the use of the third part. Based on this, we would like to organize the Issue of the fourth part. The results show that the data maintenance is necessary first, and that algorithms other than language models are suitable for such tasks, or that combinations are good. As a result, we would like to organize the necessity of the law data maintenance and the analytical and technical development.

The image on the right is from the perspective of validation, but I will omit it because it is a repetition of the content.

Let's move on to the next page. Pages 20 to 22 list the tasks that were actually used experimentally for your reference. For example, this page is a summary task, but the outline of the law proposal on the left is shown in the summary on the right, and it can be used to create materials.

On page 21, the task of idea enumeration is, in short, the task of brest. We are experimenting with a simple prompt, but I think that brest and idea generation are tasks that are often taken up as uses of ChatGPT.

On page 22, I write an example of a plausible error that is often mentioned as a Issue of a large-scale language model. The prompt on the left asks the specific content of the article, but the output on the right is different from the actual article. Since the language model alone does not perform article retrieval, such output is an unavoidable result, but on the other hand, a method to compensate for the weakness of the language model by utilizing an external API is also being studied.

On the left side, we are experimenting with such a method. Here, we are using the actual law API that is open to the public. In this way, it was combined with the processing in the language model on the right side, and we were able to obtain the actual text. This example was realized because there were the law API and the law database, but I believe that we can obtain experimental results that confirm the importance of the API and the database in utilizing such advanced services and technologies.

I have mentioned several experimental examples above, but I would like to actually make a validation and consider the ideal way of utilization, such as sorting out what is appropriate for legal affairs, what can be realized immediately, and what needs to be considered in the medium to long term.

The above is about the short-term validation of legal affairs assistance using AI, etc. Finally, on page 24, I would like to explain the draft of future policies related to the legal affairs digitalization.

My first question is that we would like to steadily conduct surveys and demonstrations on the development and utilization of legal affairs digitalization and law data.

My second question is that based on the Digital Legislation Roadmap, we would like to consider expanding the scope of law and other data to include past data and notification data, as well as expanding the functions of the API.

My third and final question is that in the above-mentioned study, we will also consider using validation to support legal affairs using AI and other technologies.

With the above as our future policy, we would like to steadily advance our deliberations.
That's all for the explanation of Material 2.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Then, I would like to take about 15 minutes for a question and answer session. If you have any questions or opinions, please let me know by raising your hand. Thank you.

Then, Mr. Tsunoda, please.

Member: , I would like to make a brief comment on this matter. official gazettes was just a topic of conversation earlier, so when law was published in official gazettes, a summary called "law Overview" was attached, and since it already exists, I think we can testing the summary result of the generative AI by comparing it with it, and there is quite excellent correct answer data, so I thought it would be good to use such a thing at this time.
That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Mr. Yamauchi, do you have any comments?

Secretariat (Yamauchi): Thank you for your visit to Thank you for your point. I would like to try it. Thank you very much.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister Then, I would like to ask you, Mr. Fujiwara.

FUJIWARA Member: I'm FUJIWARA.

I have only one point. As you pointed out, in the end, unless it is linked to an external database using an API or the like, aside from the summary, I have the impression that ChatGPT and the like are not that useful alone in this field, and I think it is quite important what it is combined with. When I think back on what I have told you so far, one of the points that you seem to be struggling with the most is the search for examples, so I don't have an idea of what and how to make it read, but it could be very useful if it could be used around the search for examples, so I personally thought it would be good to try that.

That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Next, Mr. Yasuno, please.

Yasuno Member: Just one point for .

As Mr. Yamauchi said, it is very significant to explore how LLM can be used, and I think it is wonderful that it is moved quickly.

However, one thing that I think is difficult is that technology in this field is advancing very rapidly, and we have come from a place where there was no ChatGPT six months ago, and I believe that various new models and new methods will emerge in the next six months. Therefore, although validation has been able to achieve this much and not be able to achieve this much at this point, when it is announced as the result of validation, I believe that there will be new validation. Therefore, I believe that the conclusion will be what it was at this point with some reservations, and I was listening to validation's approach while thinking that it is quite difficult to work on what continues to develop.

However, I think it is very significant, so I would like you to proceed with it.

That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

If you have any comments from Yamauchi-san, please do so.

Secretariat (Yamauchi): Thank you for your visit to .

It is true that progress in this field is quite rapid, and it is quite difficult to catch up. However, from the perspective of the digitalization of legal affairs, for example, utilization of API is important for advanced technology utilization, for example, and I believe that it will lead to the output of steadily working on it. Even if technology progresses rapidly, I believe that it will also lead to the utilization of API when testing such technology in the future, so I would like to consider and compile a final report that will lead to that.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Do you have any other questions or comments?

Mr. Yoneda, may I speak to you?

Yoneda Member: I have a lot of interest in this field, and I have great expectations for it in the future. But our mission or goal is to focus on legal affairs. In the first place, legal terms have been controlled in a normative form that is different from ordinary natural languages. In that sense, it may be more efficient to create an original database by aggregating the law and rules that have been created so far, and the data on the outline that Professor Tsunoda mentioned, than natural languages. Therefore, I think that it will be possible to produce good output or social implementation if you can compare and advance from a wider perspective.

On the contrary, I have the impression that there is a flow of mobilizing natural languages in terms of breaking down the fact that law terminology is too difficult.

That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Next, Mr. Watanabe, may I have your attention?

Watanabe, Member: Thank you, Thank you very much. Once again, I think this is a very good initiative.
As for my request, I would like to tell you a little bit about the fact that this will be an extremely great asset not only for the government and administrative officers, but also for private sector.

For example, when Airbnb, to which I am currently affiliated, is discussing various matters related to the Housing and Accommodation Businesses Act of 2017, for example, if private sector could suggest that you might consider an idea for a better article, I was a little embarrassed as an attorney. We, as lawyers, can read the article, but unlike the administrative officer here, we have not been trained to create the article, and I think we are in a situation where we cannot create the article that follows a clean and proper example.

Therefore, I would like to express my expectations for the future. In the future, this is clearly an administrative asset, but in the end, for example, it would be good if various people and various multi-stakeholders could use this article when they bring ideas to the table.
.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

If Mr. Yamauchi has any comments on the comments I received, please let me know.

Secretariat (Yamauchi): Thank you for your visit to .

I would like to respond to Mr. Watanabe's point. We, as administrators, have always considered the idea of creating a text when an idea emerges. However, I believe that it is worth considering the possibility of reflecting the ideas of the people of private sector in the text.

On the other hand, from the perspective of AI utilization, I think it is necessary to carefully consider whether or not we can immediately create a text using this language model. As Mr. Yoneda pointed out earlier, for example, law terminology is quite strict, and we have confirmed that, for example, natural-language output does not output well. In this regard, for example, we have proposed a digital legislation roadmap, but in addition to such natural-language efforts, I believe that it is necessary to development technologies that can appropriately process law terminology, for example, and to create an environment in which various law related data can be analyzed in order to development them. Therefore, I would like to conduct research and consider the necessary infrastructure development so that use cases, which you mentioned, can be realized in the future.
That's all.

Watanabe, Member: Thank you, .

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

How about others?

Member Yasuno, please.

Yasuno Member: Just one point for Quick.
I believe that the fact that natural language and legal language are completely different, which Professor Yoneda mentioned earlier, is a quite important point. In fact, the current LLM can handle machine language quite well, so in a sense, I think it has begun to be confirmed that it can handle quite a lot of things other than human natural language.

On the other hand, legal language is not included in training at all compared to code, so I personally doubt how well it can be handled by ChatGPT at this point, but I think there is a possibility that such language will be strengthened in the future.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Mr. Yoneda, may I speak to you?

Yoneda Member: I'm sorry to bother you again.

There was a comment that it is a survey and research on the current status and future of digital legislation that we would like to connect education institutions and research with the first POC and experiments using the current LLM. I think it is very important how we will do this.

Currently, there are no university research institutes that are actually doing this at the level we need, so I am worried that there will be no human resources to take over this part. There is a problem that bureaucrats are being transferred more and more, and that knowledge is not being accumulated because the person in charge is changing. In that sense, I feel that unless we quickly think about setting up a base somewhere in the university or creating and developing a education program at several places, we will probably not be able to acquire people and keep up with it. Now, I am very excited about it, so I feel that I will be able to do something about it, but after a while, I feel that I will face the problem of not having a successor, so I don't think I can fully discuss it here, but I hope that you will be aware of it and work on it as a whole.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister That's right. Thank you very much. I would like to make efforts while taking into account the development and succession of human resources, as well as the establishment of a network inside and outside the government.

Do you have any comments from Mr. Yamauchi?

Secretariat (Yamauchi): Thank you for your visit to .

In addition to what I have just mentioned, in considering the utilization of law Database and digital technologies, from the perspective of having studied computer science, I myself thought it was interesting while working in law and worked on the analysis of law Database. I would be happy if more people would be interested in law Database and API by disclosing it. Based on this, for example, we are considering a two sided design and trial use of API based on users, and I would like to work on fostering interest including external people.

That's all.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister .

Do you have any other questions or comments?

If you don't mind, based on the various points you pointed out and the encouragement you gave me today, I have set a very challenging Issue, but I will continue to work with enthusiasm this fiscal year.

Then, if you don't mind, I would like to have a closing address by Senior Vice-Minister for Digital Okushi, who is the chairman of this study team. Senior Vice-Minister, nice to meet you.

Parliamentary Senior Vice-Minister Okushi: I am OGUSHI Masaki, Parliamentary Senior Vice-Minister for Digital . Thank you very much for your active discussion today.

Today, I would like to thank all the members for their very valuable comments on our efforts to digitize review of digital legislation and official gazettes, the survey and demonstration of the development and utilization of digitalization and law data for legal affairs, and the utilization of AI.

Going forward, we will steadily implement review of digital legislation in Digital Agency based on the results of inspections by the Cabinet Office and each Ministry. In addition, regarding the digitization of official gazettes, we would like the Cabinet Office, which has been participating as an observer since today, to proceed with consideration of the bill based on today's discussions. Digital Agency will provide the necessary cooperation.

In addition, regarding the investigation and demonstration of the development and utilization of digitalization and law data for legal affairs, we are finally demonstrating the contents discussed by the validation Team over a long period of time. Once again, we would like to thank you for your support so far. In the validation of this investigation and demonstration and the utilization of a large-scale language model for legal affairs, which has been a topic of conversation recently, the support of people with high expertise is essential, so we would like to ask for your continued guidance.

In addition, we would like to ask for the continued cooperation of the Cabinet Legislation Bureau, which has been participating as an observer since today.

Once again, thank you very much for your time today.

Secretariat (Mr. Yamaguchi): Thank you, Parliamentary Senior Vice-Minister

If you don't mind, I'd like to conclude today's proceedings.

With regard to today's proceedings, if you do not have any objection, I would like to prepare the minutes, have them checked by all of you, and disclose them, as well as the materials.

With that being said, I would like to conclude today's meeting. Thank you for taking the time to join us today.