IGF 2023 Town Hall #165 Exploring the Risks and Rewards of Generative AI

Thursday, 12th October, 2023 (05:45 UTC) - Thursday, 12th October, 2023 (06:45 UTC)
WS 3 – Annex Hall 2

Chat GPT, Generative AI, and Machine Learning

Other - 60 Min
Format description: Town Hall meeting; table for speakers facing seating for participants and screen of remote participants.


This Town Hall delves into the multifaceted aspects of generative AI, analyzing both the risks and benefits it brings to the table, with a special emphasis on enhancing safety and security. We will start with a fictional scenario to help generate debate about safer concerns and AI. The moderated discussion to follow will crowdsource concerns regarding the potential dominance of super intelligent AI, while also highlighting the positive impacts that AI can have across various domains. The moderator, rapporteur, and online moderator will delve into how generative AI can bolster security measures, shape surveillance practices, foster online safety, contribute to the public good, and elevate user experiences. It is crucial to acknowledge the potential misuse of generative AI, as it can be employed to create harmful or inappropriate content. As we navigate this realm, the session will underscore the significance of an exploratory approach that balances caution with innovation, avoiding undue moral panics and highlighting real security concerns.

We would like to crowdsource participant concerns about safety related to generative AI. This aspect is not well accounted for in public debate, and differ, based on the audience and the application of the technology. To help get reactions, we propose running a security exercise scenario for participants, to highlight issues and spark debate. Our onsite moderator facilitates Safer Internet Day for the US. He's been skilled at integrating virtual and in-person audiences for years. Our online moderator will be responsible for facilitating onsite and online interactions. He's run several conference and is bilingual, so we hope to get more participating. Our onsite rapporteur has run World Cafes at RightsCon and AI safety training exercises for the Christchurch call, so she will be developing a scenario for participants to respond to. It will be made available beforehand. This was done via zoom and videoconferencing should be sufficient.


Stanford University
Larry Magid, ConnectSafely, Civil Society (U.S. facilitator of Safer Internet Day), North America Brittan Heller, Stanford University, Civil Society, North America Daniel Castano, Universidad Externado de Colombia, Civil Society, South America


Larry Magid, ConnectSafely, Civil Society (U.S. facilitator of Safer Internet Day), North America Brittan Heller, Stanford University, Civil Society, North America Daniel Castano, Universidad Externado de Colombia, Civil Society, South America

Onsite Moderator

Larry Magid

Online Moderator

Daniel Castano


Brittan Heller


3. Good Health and Well-Being
11. Sustainable Cities and Communities
16. Peace, Justice and Strong Institutions
17. Partnerships for the Goals

Targets: Our proposed panel on the risks and benefits of generative AI relates to multiple UN Sustainable Development Goals. It addresses how generative AI can contribute to good health and well-being by improving healthcare outcomes, enhancing security and safety measures in sustainable cities and communities, promoting peace, justice, and strong institutions through ethical considerations and regulatory frameworks, and fostering partnerships among stakeholders for responsible AI practices in alignment with the SDGs.

Key Takeaways (* deadline 2 hours after session)

In response to challenges emerging from GAI, all stakeholders should prioritize responsible AI development, implement robust cybersecurity measures, establish clear accountability mechanisms, and promote international collaboration. Addressing these concerns is crucial to mitigate the potential issues and risks associated with AI in national security and information dissemination.

The AI digital divide remains a persistent and pressing challenge for GAI, as it may amplify inequities between different global populations.

Call to Action (* deadline 2 hours after session)

Remain cautious of heuristics used to determine authentic content, as AI technology improves. These will quickly become outdated.

Focus on encouraging creativity and critical thinking, especially for youth and in civic education. This will help ensure that GAI has a more positive impact on global societies.

Session Report (* deadline 26 October) - click on the ? symbol for instructions

Report on the 2023 IGF Town Hall Session: Exploring the Risks and Rewards of Generative AI


The 18th annual United Nations Internet Governance Forum (IGF) held in Kyoto, Japan, featured a diverse range of sessions, including one Town Hall sesson entitled “Exploring the Risks and Rewards of Generative AI.” The session addressed the growing interest and concerns related to Generative AI (GAI) and its impact on content creation, ethics, education, cybersecurity, and global equity. The panelists in this session provided valuable insights into the challenges and opportunities associated with GAI.

Panelists and Moderator

  1. Daniel Castaño - Law Professor from the Universidad Externado de Colombia.
  2. Zoe Darme - Senior Manager at Google.
  3. Brittan Heller - Senior Fellow at the Atlantic Council and Affiliate at the Stanford Cyber Policy Center.
  4. Larry Magid – Panel moderator; Founder and CEO of ConnectSafely
  5. Janice Richardson - Senior Advisor to Insight, a European internet safety education organization.

Key Discussions and Concerns

  1. Ethical Biases and Trust in GAI:

Brittan Heller highlighted the risks of conversational AI systems, such as chatbots, inadvertently propagating moral and ethical biases, which can erode trust and lead to public dissatisfaction. She expressed concerns about the lack of accountability in environments where authenticating content or indicating its provenance is challenging, potentially leading to misinformation or bias. Additionally, she warned about malware, cybersecurity risks, foreign interference, deep fakes, and their profound economic and societal consequences.

  1. AI in Education:

Janice Richardson discussed both the opportunities and challenges of AI in K-12 education. She pointed out the potential for AI to enhance personalized learning, overcome disabilities, provide data-based feedback, reduce skill gaps, and automate repetitive tasks. However, she also raised concerns about the geopolitical impact of AI, its potential to exacerbate systemic racism and inequality, and the risk of bias in AI algorithms negatively impacting marginalized communities.

  1. Validation and Authenticity Challenges:

Zoe Darme emphasized the challenges of validating content, noting that common heuristic shortcuts are not always reliable. She highlighted how generative AI can create authentic-looking messages, making it difficult for individuals to discern accuracy and authenticity. She also mentioned the difficulty in identifying content as AI-generated, as both human-created and AI-generated content can be inaccurate or misleading.

  1. Global Equity and Digital Divide:

Daniel Castaño discussed how GAI could affect the digital divide in global majority countries. He pointed out the danger of AI further increasing inequality both within and between countries. Algorithms may favor wealthy nations while negatively impacting people in developing countries.

  1. Historical Perspective:

Larry Magid acknowledged the concerns and excitement over GAI and placed them in a historical context. While recognizing that GAI will have both positive and negative consequences, they stressed the need to avoid moral panics and exaggerated fears. He emphasized that the impact of GAI on society remains uncertain, but it is unlikely to pose an
“extinction-level threat” comparable to nuclear war, climate change, or pandemics.


“Exploring the Risks and Rewards of Generative AI” provided a comprehensive overview of the challenges and opportunities associated with GAI. The panelists’ insights and audience questions highlighted the ethical, educational, validation, and equity-related concerns while also acknowledging the potential benefits of GAI. The discussion underscored the importance of responsible development and use of generative AI to maximize its positive impact and mitigate potential risks. It remains crucial for policymakers, industry experts, and researchers to work together to shape the future of AI in a way that benefits society as a whole while addressing these concerns.