人工知能標準化国際シンポジウム:Trustworthy AI 実現へ向けたAI標準化International Symposium on AI Standards: towards trustworthy AI through standards

2024年7月3日 (水) ~ 5日(金)
3 July (Wed) – 5 July (Fri), 2024
日本科学未来館 7F イノベーションホール
Innovation hall, 7th floor, The National Museum of Emerging Science and Innovation
〒135-0064 東京都江東区青海2-3-6 日本科学未来館
2-3-6 Aomi, Koto-ku, Tokyo, 135-0064 Japan
(10時より前の入口は職員通用口をお使いください, entrance before 10:00 am: 35.619351N, 139.776093E)
主催:産総研人工知能標準化委員会(経済産業省の委託事業により開催)
Organizer: AI standardization committee, National Institute of Advanced Industrial Science and Technology (AIST),
entrusted by the Ministry of Economy, Trade and Industry (METI)

参加登録は以下のFORMSからお申し込みください。先着順での受付となります。


https://forms.office.com/r/RaWsSPUHfk

趣旨

AIは社会へもたらす益もリスクも極めて大きい。このため、技術の影響が社会に適切に現れるような仕組みづくりが重要となる。そこで各国は国際標準を含むルール作りをAI政策と緊密に結びつけ、技術開発と並行して行うことでAIの社会受容の地盤硬め(リスクを低減させ,安心できる社会を実現する)を進めようとしている。この標準、政策、技術開発という三つ巴の構造は古典的な標準化、「市場へ商品が出てから業界とりまとめのために行う活動(MPEG,デジタルTV等)」とは全く異なる。国際標準策定に向けた合意形成もこの三つ巴構造を意識し、技術に加えてルール形成担当者を、そして政策担当者を十分に巻き込んで行う必要がある。

日本はこれまで、この構造を意識しながらデータ品質や機能安全、ライフサイクル等のISO/IEC標準をエディタやコンビーナを出し開発してきた。そして今、ISO/IECにヒューマン・マシン・チーミング (HMT) という、人とAIとの関係を統一的に整理する概念を提案し、またISO/IEC 42001 (AIマネジメントシステム (AIMS) 標準) に代表されるAI認証についてのエコシステムづくりを、海外を行脚し巻き込みつつ行っている。

本シンポジウムは首記の三つ巴構造を、その現場で政策や標準化を担うグローバルなリーダー達からのご講演により示す。そして彼らに対し、これまで日本が注力してきた標準を、そしてHMTとAIMSを位置づけることで日本が考える今後のAI標準化の方向性を示す。これによりAIの国際標準化の、そして日本の標準化活動に対する各国の政策担当者やルール形成担当者、標準化担当者の理解を深め、HMTとAIMSの、そして将来のAI標準作りの礎とする。

Objective

AI brings both extremely large benefits and risks to society, which makes it critical to establish a mechanism to ensure that the impact of the technology is appropriately reflected in society. Countries are therefore closely linking rule-making, including international standards, with AI policy, and are working in parallel with technological development to harden the ground for social acceptance of AI (i.e. to reduce risks and realise a secure society). This three-way structure of standards, policy and technological development is completely different from classical standardization, which is 'activities carried out for industry coordination after a product has been released to the market (MPEG, digital TV, etc.)'. Consensus building for the formulation of international standards also needs to be aware of this three-way structure and fully involve those in charge of rule-making as well as technology, and those in charge of policy.

Japan has so far developed ISO/IEC standards for data quality, functional safety, lifecycle, and others with an awareness of this structure by sending out editors and conveners. Now, Japan has proposed the concept of Human-Machine Teaming (HMT) to the ISO/IEC to organize the relationship between humans and AI in a unified manner, and is involved in the creation of an ecosystem for AI certification, represented by ISO/IEC 42001 (AI Management System (AIMS) standard), by visiting and involving other countries.

This symposium will present the above three-way structure through lectures by global leaders in charge of policy and standardization. It will also present the direction of future AI standardization in Japan by positioning HMT and AIMS alongside the standards that Japan has been focusing on. This will deepen the understanding of policy makers, rule makers and standardizers of AI international standardization and Japan's standardization activities, and will form the basis for HMT and AIMS and future AI standard-making.

Program

3rd July 2024 (Wednesday) 09:00 – 12:20: AI Global Now (1) Chair: Dr. Roy Sugimura (AIST)

08:30

Registration open

09:00 – 09:10

Greetings from distinguished guests

09:10 – 09:15

Why now? Another trialogue

Prof. Dr. Roy Sugimura (Supervisory Innovation Coordinator at National Research Institute for Advanced Industrial Science and Technology (AIST)) Chair, AI standardization committee, AIST
Chair, ISO/IEC JTC1/SC42 Japan mirror committee
09:20 – 10:00

SC42 in Worldly context (tentative)(remote)

Mr. Wael William Diab Chair, ISO/IEC JTC1/SC42
10:00 – 10:40

US Approach (tentative) (remote)

Ms. Elham Tabassi Chief AI Advisor, NIST Associate Director for Emerging Technologies at Information Technology Laboratory, NIST
10:40 – 11:00

Break

11:00 – 11:40

Beyond AI standards and policy: Digital trust ecosystems from a European perspective

Dr. Sebastian Hallensleben Sebastian Hallensleben is the Chair of CEN-CENELEC JTC 21 where European AI standards to underpin EU regulation are being developed, a member of the Expert Advisory Board of the EU StandICT programme and Chair of the Trusted Information working group. He co-chairs AI risk and accountability in OECD ONE.AI and has roles in AI committees in the Council of Europe and UNESCO. Sebastian Hallensleben heads Digitalisation and Artificial Intelligence at VDE Association for Electrical, Electronic and Information Technologies. He focusses in particular on AI ethics, on handling the impact of generative AI, building privacy-preserving trust infrastructures as well as characterising AI quality.
11:40 – 12:20

The future of certification for digital services: Enabling trust through new methods of conformity assessment (remote)

Mr. Olivier Colas Senior Director International Standards at Microsoft

3rd July 2024 (Wednesday) 14:00 – 17:00: AI Global Now (2) Chair: Prof. Takashi Egawa (AIST / Keio-U)

14:00 – 14:30

Reality of AI standardization and Canada’s approach (tentative)

Mr. Paul Cotton (Founder and President, PBC & Associates) Convenor, ISO/IEC JTC 1/SC 42/WG 1 and Co-convenor, SC 42/AHG 7
Vice- Chair, ISO/IEC JTC 1/SC 42 Canada mirror committee
14:30 – 15:00

European approach (tentative)

Mr. Patrick Bezombes Vice-chair, CEN-CENELEC/JTC 21
Convenor, CEN-CENELEC/JTC 21/Strategic Advisory Group
Chair, AFNOR standardization committee on AI
15:00 – 15:30

Human-Machine Teaming in Australia

Mr. Harm Ellens (Director, Virtual Ink Australia) Convenor ISO/IEC JTC 1/SC 42 Joint Advisory Group with SC 39 on AI and sustainability
Project Editor ISO/IEC TR 20226, Environmental sustainability aspects of AI systems
Co-convenor SC 42/WG 3 Trustworthiness Roadmapping adhoc
Australian observer at ISO/IEC JTC21 Artificial Intelligence
Standards Australia IT-43 Artificial Intelligence mirror committee WG4 convenor
15:30 – 16:00

Break

16:00 – 16:30

Uganda’s approach (tentative)

Ms. Barbara Ritah Atim (Standards Officer, Uganda National Bureau of Standards) Experienced standards developer and promoter with expertise in Machine Learning, Deep Learning, Data Science and Engineering.
Project editor of ISO/IEC TR 42109 Use cases of human-machine teaming
16:30 – 17:00

Singapore’s Approach to AI Governance (remote)

Mr. Benjamin Chua Senior Manager (AI Governance), Infocomm Media Development Authority, Singapore

3rd July 2024 (Wednesday) 18:00 – 20:30 Banquet

18:00 – 20:30

Katsuo-Tamago Shokudo カツオたまご食堂 (35.617782N, 139.778439E)

4th July 2024 (Thursday) 09:00 – 12:30: Human-Machine Teaming Chair: Dr. Fumi Maruyama (Invited Senior Researcher, AIST)

08:30

Registration open

09:00 – 09:30

Human-Machine Teaming: Overview

Dr. Fumihiro Maruyama (Invited Senior Researcher, AIST) Convenor of ISO/IEC JTC 1/SC 42/WG 4 (Use cases and applications)
Secretary of ISO/IEC JTC 1/SC 42 Japan Mirror Committee
Special advisor on AI education, Chuo Computer & Communication College Group
09:30 – 10:00

A Human-Machine Teaming Approach to Beneficial AI Systems (remote)

Prof. Dr. Stephen Kwan (Professor Emeritus, San Jose State University) Member of the American National Standards Institute (ANSI) Board of Directors and project editor of ISO/IEC TR 5259-6 (Visualization framework for data quality) and ISO/IEC TR 21221 (Beneficial AI systems)
10:00 – 10:30

Standardizing Human-Machine Teaming

Dr. Yuchang Cheng (Senior Research Manager, Fujitsu) Project editor of ISO/IEC TR 24030.1:2021 Use cases (ed.1), TR 24030.2:2024 Use cases (ed.2), ISO/IEC 5338:2023 AI system life cycle processes and ISO/IEC PWI Human-machine teaming
Ms. Barbara Ritah Atim (Standards Officer, Uganda National Bureau of Standards) Experienced standards developer and promoter with expertise in Machine Learning, Deep Learning, Data Science and Engineering.
Project editor of ISO/IEC TR 42109 Use cases of human-machine teaming
10:30 – 11:00

Break

11:00 – 11:20

Human-Machine Teaming in the Field of Learning, Education and Training

Dr. Hdr. Frederic Andres (Associate Professor, National Institute of Informatics) Convenor of ISO/IEC JTC 1/SC 36/AHG 5 (Blockchain in Education) and project editor of ISO/IEC 20016-1:2014 Language accessibility and human interface equivalencies (HIEs) in e-learning applications — Part 1: Framework and reference model for semantic interoperability and ISO/IEC 29187-1:2013 Identification of privacy protection requirements pertaining to learning, education and training (LET)
11:20 – 11:40

Human-Machine Co-Evolution System Guidebook

Mr. Akihisa Morikawa (Director, Imaginary Corporation) Corporate Officer of Witz Corporation, Service Design Business Area
Project Leader of HMCES (Human-Machine Co-Evolution Systems) project aiming to promote human-machine co-evolution and prevent/mitigate new risks
11:40 – 12:30

Wrap-up Panel

Mr. Harm Ellens (Director, Virtual Ink Australia) Convenor ISO/IEC JTC 1/SC 42 Joint Advisory Group with SC 39 on AI and sustainability
Project Editor ISO/IEC TR 20226 (Environmental sustainability aspects of AI systems)
Co-convenor SC 42/WG 3 (Trustworthiness Roadmapping adhoc)
Australian observer at ISO/IEC JTC21 (Artificial Intelligence)
Standards Australia IT-43 Artificial Intelligence mirror committee WG4 convenor

4th July 2024 (Thursday) 14:00 – 14:30: AI Global now (3)

14:00 – 14:30

UK approach (tentative)

Prof. Dr. David Leslie Director of Ethics and Responsible Innovation Research, The Alan Turing Institute
Professor, Digital Environment Research Institute, Queen Mary University of London

4th July 2024 (Thursday) 14:30 – 16:55: AI Management System Chair: Professor Emeritus, Yo Harada (Institute of Information Security)

14:30 – 14:45

Introduction to AIMS standards developments

Prof. Yonosuke Harada, Emeritus Professor Institute of Information Security
Convener, ISO/IEC JTC 1/SC 42/JWG 1
14:45 – 15:15

Standardization of AIMS in ISO/IEC JTC 1/SC 42/WG 1 (tentative)

Mr. Paul Cotton (Founder and President, PBC & Associates) Convenor, ISO/IEC JTC 1/SC 42/WG 1 and Co-convenor, SC 42/AHG 7
Vice- Chair, ISO/IEC JTC 1/SC 42 Canada mirror committee
15:15 – 15:45

Break

15:45 – 16:05

Introduction to ISO/IEC 42001 (tentative)

Ms. Marta Janczarski (National Standards Officer, Microsoft Ireland) Project Editor, ISO/IEC 42001:2023 (AI -- Management System)
Chair, Head of Delegation for National Standards Authority Ireland, Cybersecurity and Privacy
16:05 – 16:25

Details of ISO/IEC 42001 (tentative)

Dr. Hiroki Takamura (JQA) Chair, JIS Q.42001 development committee
16:25 – 16:55

Making AIMS certifiable – an overview of ISO/IEC 42006

Ms. Susanne Kuch (Digital policy officer of the German Accreditation Body (DAkkS)) Project Editor, ISO/IEC 42006, Requirements for bodies providing audit and certification of artificial intelligence management systems

5th July 2024 (Friday) 09:00 – 12:00: Japan’s commitment to ISO/IEC JTC 1/SC 42 Chair: Dr. Yuchang Cheng (Senior Research Manager, Fujitsu, Japan)

08:30

Registration open

09:00 – 09:10

Japan’s commitment to ISO/IEC JTC1/SC42

Prof. Dr. Roy Sugimura (Supervisory Innovation Coordinator at National Research Institute for Advanced Industrial Science and Technology (AIST))Chair, AI standardization committee, AIST Chair, ISO/IEC JTC 1/SC 42 Japan mirror committee
09:10 – 09:40

Standards relative to AI use cases and applications (tentative)

Mr. Nobuhiro Hosokawa (General Manager, IBM, Japan) Technical consultant, IBM Consulting, Manager, Financial Business Solutions
09:40 – 10:10

Data quality standards for analytics and machine learning (5259-X)

Dr. Kyoungsook Kim Deputy Director of the Artificial Intelligence Research Center (AIRC) of AIST,
Project Editor of ISO/IEC 5259-2 (Data quality measures)
10:10 – 10:45

Break

10:45 – 11:00

Functional Safety, TR 5469 and TS 22440 (tentative)

Dr. Yutaka Oiwa Deputy Director, Digital Architecture Research Center, National Institute of Advanced Industrial Science and Technology (AIST)
11:00 – 11:15

Standardization of human oversight

Prof. Takashi Egawa (AIST/Keio-U) Project editor of ISO/IEC 42105 (Guidance for human oversight of AI systems), ISO/IEC TR 5469:2024 (Functional safety and AI systems), Secretary of IEEE 7001-2021 (Transparency of Autonomous Systems)
11:15 – 11:30

New facts from AIMS feasibility studies for AIMS and proposals for constellations of standards around ISO/IEC 42001

Prof. Yonosuke Harada, Emeritus Professor Institute of Information Security
11:30 – 11:45

Application of AI technologies in health informatics (tentative)

Prof. Dr. Shusaku Tsumoto Professor of Medical Informatics, Department of Medical Informatics, Shimane University, Faculty of Medicine President, Japanese Society of Artificial Intelligence (JSAI)
11:45 – 12:00

Making SC42 standards into JIS (tentative)

Dr. Roy Sugimura Chair, JIS X.22989 development committee

Mr. Hiroyuki Ogura Chair, JIS Q.38507 development committee

Dr. Hiroki Takamura (JQA) Chair, JIS Q.42001 development committee

5th July 2024 (Friday) 14:30 – 18:00 Future Direction (invitation only, Chatham House Rule)
Coordinator : Prof. Dr. Arisa Ema (U-Tokyo)

14:30 – 18:00
Invited Experts (presenters of this symposium), AIST AI Standardization Committee experts, experts on AI standards, and invited participants.
Based on the presentations at the symposium, participants discuss current conundrum in the standardization of AI, e.g., Global & Regional (Regulations, Guidelines), Vertical and Horizontal (Industries), and What is the real change?.

会場案内図

https://www.miraikan.jst.go.jp/visit/facility-rental/

Venue: innovation hall (No. 8)

事務局(株式会社ヒップ内)

TEL:03-3370-2411(土・日・祝日を除く10:00~17:30)/ FAX:03-3370-2017

E-MAIL:symposium@hip-ltd.co.jp