IGF 2023 Networking Session #109 International Cooperation for AI & Digital Governance

SESSION SUMMARY

• A Human Rights Approach to Digital Governance (Matthew Liao, NYU)
  
  • "5W1H " framework for AI regulation, focusing on several key questions: What should be regulated? Why should AI be regulated? Who should regulate AI? When should regulation begin in the technology life cycle? Where should regulation occur? How should regulation be enacted?
  • AI regulation should prioritize the protection and promotion of human rights, encompassing everything that could impact these rights and everyone including companies, researchers, governments, universities, and the public, has a responsibility to engage proactively in the regulation process to overcome challenges in enforceability.

• AI and Cyber Physical Systems Policy Lab (Dasom Lee, KAIST)
  
  • AI&CPS lab focuses on the integration of AI into infrastructure to promote environmental sustainability. The lab investigates areas like energy transition, smart grids, renewable energy technologies, transportation (including automated vehicles and drones), and data centers. She emphasized the interconnectedness of these fields, highlighting the importance of a harmonious approach to infrastructure development.
  • One of the on-going projects, ‘Privacy and Culture Project’ shows that privacy is contextualized in different geographical regions based on culture and history, which calls for an international cooperation of experts and academics for tackling future challenges.

• Challenges and Opportunities of Digital Governance for Development in the Developing Countries (Atshushi Yamanaka, JICA)
  
  • Digital governance and technologies offer opportunities for innovation and AI can create new products and services that contribute to socio-economic development. “Reverse innovations”, that emerge from developing countries can impact the global landscape, which can be fostered by enhancing digital public goods and infrastructure to ameliorate the digital divide.
  • Digital governance requires a multi-stakeholder approach inviting stakeholders in policy-making processes, and more platforms like IGF if needed in the future, especially to overcome new challenges and to invite developing countries need to take a more significant role in rule-making and framework development.

• Conversational Agents for Digital Inclusion (Rafik Hadfi, Kyoto University)
  
  • Digital inclusion goes beyond access to technology and also encompasses concepts like equity, self-realization, and autonomy, and ultimately aims to disadvantaged individuals with access to ICT technology and, in doing so, promote equity.
  • A case study in Afghanistan showed how the AI system can enhance several aspects of the online debates, including increasing the diversity of contributions from women, reducing inhibitions, and encouraging more ideation on local problems.

• Democratising AI for a thriving region (Liming Zhu, University of New South Wales)
  
  • In 2019, Australia introduced AI ethics principles that prioritize human-centered values such as pluralism, fairness, and inclusiveness, as well as quality attributes specific to AI, including privacy, security, transparency, and accountability.
  • As part of this comprehensive approach to AI governance, Australia has introduced measures such as responsible AI risk assessments, question banks, and AI risk registries to guide organizations in identifying and mitigating potential risks associated with their AI applications. This approach aims to ensure that AI benefits are leveraged responsibly and ethically, with consideration for social and environmental well-being.

• Towards Hyperdemocracy: Discussion and Consensus among People and Machines (Takayuki Ito, Kyoto University)
  
  • The Hyperdemocracy project aims to utilize artificial intelligence (AI) to address contemporary social challenges such as fake news, digital manipulation, and echo chambers, with a transition in 2015 to develop a system that leverages AI agents to support group interactions and crowd-scale discussions online.
  • A case study was conducted in Afghanistan to collect public opinions during the withdrawal of American troops in August 2021, with the AI analyzing the types and characteristics of these opinions. The project is currently developing a next-generation AI-based competition support system to address what they refer to as the "hypodemocracy problem."

• International Development and AI: Three Snapshots (Seung Hyun Kim, KAIST)
  
  • Unequal opportunities can arise from technological advancements and introduction of new technologies can create unexpected problems arising from existing social, economic disparities.
  • Fragmentation in governmental ICT systems can occur due to the involvement of different financing institutions and service providers.
  • Technology sovereignty is a critical concern, and the dynamic between efforts to keep nation's security with technological innovation will become increasingly difficult.