Beneficial AI Tokyo
Artificial Intelligence (AI) promises to be one of the most transformative technologies in human history. From academia, industry, and the policy world, a growing number of organizations are now working to ensure that it is safe and beneficial. For the moment, most of these initiatives are Western, mainly from Anglophone countries. These voices need to be joined by others. The task of ensuring that the development of AI benefits all of humanity calls for a genuinely global community, and a forum for voices from all over the world.
How can we build an inclusive and effective global community? Beneficial AI Tokyo was held in association with a symposium on AI and Society on October 12, 2017 to explore this challenge, with a particular focus on Japan.
The participating academic groups and NGOs, corporate groups, and others presented their thoughts on effective ways to build cooperation for beneficial AI, and the participants discussed topics such as:
- Corporate voices for beneficial AI – What can corporations do to encourage cooperation on the challenges of ensuring that AI is beneficial?
- Basic research and beneficial AI – What can scientists doing basic research on AI and related subjects do to help ensure that AI is beneficial?
- Near-term choice points for long-term risks and benefits – Are there choices about the direction of development of AI in the near-term which may have long-term consequences?
- Building Japan’s beneficial AI community – What are the next steps to encourage cooperation for beneficial AI in Japan?
- Working across borders for beneficial AI – How can we encourage international cooperation and collaboration for beneficial AI?
- Technical AI safety – What are the technical challenges of ensuring that AI is safe and reliable and aligned with human values?
- Technical issues of bias and privacy – How can we ensure fairness and respect for privacy in the application of machine learning for making decisions about people?
- Regulation for beneficial AI? – Do we need regulation for the development of AI? If so, should it be at a national or international level?
- AI ethics: who gets a voice? – AI is likely to affect the lives of everybody; who decides how it should be applied?
- Avoiding an AI arms race – Are there dangers if the development becomes an ‘arms race’ between global powers? If so, how can they be avoided?
- Visions of a positive AI future – If the development of AI goes well over the next 50+ years, what sort of world will our grandchildren be living in? What are the long-term benefits?
- AI for the bottom billion – The world’s poorest people often have least access to new technologies. What can be done to ensure that they get the benefits of AI?
- Community-building for beneficial AI – What can be done at a grass-roots level to encourage widespread interest and involvement in the challenges of beneficial AI?
The participants expressed commitment to working towards AI for good, and endorsed a Conference Statement “Cooperation for Beneficial AI”.
|09:00 – 09:30||Arrival & Registration|
|09:30 – 09:45||Welcome & Introduction by Huw Price (CFI/CSER) & Ryota Kanai (Araya)|
|09:45 – 11:15||NGO panel session|
Moderator: Arisa Ema (University of Tokyo)
Speakers: John C. Havens (IEEE Global AI Ethics Initiative),
Viktoriya Krakovna (Future of Life Institute),
Yasuo Kuniyoshi (AI Center), Seán Ó hÉigeartaigh (CSER/CFI),
Bing Song (Berggruen Institute China Center),
Hideaki Takeda (JSAI), Denis Therien (Element AI/CIFAR)
|11:15 – 11:30||Coffee break|
|11:30 – 13:00||Corporate panel session|
Moderator: Jaan Tallinn (CSER)
Speakers: Olga Afanasjeva (GoodAI), Jake Lucchi (Google HK),
Masaya Mori (Rakuten), Yutaka Matsuo (Japan Deep Learning Association),
Francesca Rossi (Partnership on AI), Akira Sakakibara (Microsoft Japan)
|13:00 – 14:30||Lunch|
|14:30 – 16:30||Breakout sessions|
|16:30 – 17:45||Concluding panel & discussions|
Co-chairs: Ryota Kanai (Araya), Huw Price (CFI/CSER)
Panelists: Arisa Ema (University of Tokyo), Danit Gal (Peking University),
Toyoaki Nishida (Kyoto University), Osamu Sakura (University of Tokyo),
Hideaki Shiroyama (University of Tokyo), Jaan Tallinn (CSER)
|17:45 – 18:00||Closing statement by Yasuo Kuniyoshi (AI Center) & Ryota Kanai (Araya)|
|18:00 – 20:00||Reception|
- Next Generation Artificial Intelligence Research Center of the University of Tokyo (AI Center)
- Araya, Inc.
- The Leverhulme Centre for Future of Intelligence (CFI)
- The Centre for the Study of Existential Risk (CSER)
Organizing Committee Members
- Huw Price, CFI
- Seán Ó hÉigeartaigh, CSER
- Ryota Kanai, Araya
- Ayako Fukui, Araya
- Arisa Ema, University of Tokyo
- Francesca Rossi, IBM/Partnership on AI
- John Havens, IEEE Global Initiative
- Danit Gal, Peking University
- Shahar Avin, CSER
- Martina Kunz, CFI
- Jaan Tallinn, CSER/Future of Life Institute (FLI)
- Ralf Herbrich, Amazon/Partnership on AI
- Anthony Aguirre, FLI
- Yutaka Matsuo, University of Tokyo/JSAI Ethics Committee
- Osamu Sakura, University of Tokyo