Session
Description:
The UN Universal Periodic Review (UPR) has proved its potential to transform peer pressure into tangible change many times. Mongolia adopted legislation to protect human rights defenders. Serbia rebuilt its anti-discrimination safeguards following recommendations. Dominica formally joined the torture convention mid-process. These shifts show that States act when scrutinized by peers. Yet artificial intelligence governance, though increasingly featured, often remains inconsistently enforced. This shows how algorithmic power eludes accountability architecture, especially when the mechanism is non-bidding and the authoritarianism advances. Now comes the harder test. If the UPR succeeded in transforming state behavior on torture and discrimination, what would be needed to make it equally effective in constraining the proliferation of unaccountable AI systems? The path forward hinges on various dimensions: making algorithmic harms visible; providing implementation-oriented recommendations and follow-up steps that replace generic commitments; and building capacity support for states to integrate UPR-AI assessments into domestic policy, especially in countries of the Global South. This session examines how to turn the UPR into a sustained advocacy and accountability tool for AI governance, understanding the gap between peer pressure and institutional change. Its impact depends on translating recommendations into national monitoring and follow-up mechanisms. The workshop brings together expert voices from the Civil Society Alliances for Digital Empowerment (CADE), OHCHR, and state-mandated bodies, with perspectives from both the North and the South. The IGF remains a unique convening space for these and other multistakeholders to converge and strategize on advancing AI governance. This session will surface lived experiences and institutional capacities, acknowledging that the UPR's real power isn't just about the recommendations but in what happens after, when civil society decides to hold States accountable to their own commitments.
The UN Universal Periodic Review (UPR) has proved its potential to transform peer pressure into tangible change many times. Mongolia adopted legislation to protect human rights defenders. Serbia rebuilt its anti-discrimination safeguards following recommendations. Dominica formally joined the torture convention mid-process. These shifts show that States act when scrutinized by peers. Yet artificial intelligence governance, though increasingly featured, often remains inconsistently enforced. This shows how algorithmic power eludes accountability architecture, especially when the mechanism is non-bidding and the authoritarianism advances. Now comes the harder test. If the UPR succeeded in transforming state behavior on torture and discrimination, what would be needed to make it equally effective in constraining the proliferation of unaccountable AI systems? The path forward hinges on various dimensions: making algorithmic harms visible; providing implementation-oriented recommendations and follow-up steps that replace generic commitments; and building capacity support for states to integrate UPR-AI assessments into domestic policy, especially in countries of the Global South. This session examines how to turn the UPR into a sustained advocacy and accountability tool for AI governance, understanding the gap between peer pressure and institutional change. Its impact depends on translating recommendations into national monitoring and follow-up mechanisms. The workshop brings together expert voices from the Civil Society Alliances for Digital Empowerment (CADE), OHCHR, and state-mandated bodies, with perspectives from both the North and the South. The IGF remains a unique convening space for these and other multistakeholders to converge and strategize on advancing AI governance. This session will surface lived experiences and institutional capacities, acknowledging that the UPR's real power isn't just about the recommendations but in what happens after, when civil society decides to hold States accountable to their own commitments.
Policy Question(s)
- What methodologies are useful when assessing AI human-rights impacts, including those that are most difficult to measure, such as indirect or cumulative harms?
- How can UPR AI recommendations be more specific and implementation-oriented, and avoid generic commitments that do not change domestic policy? What follow-up actions are necessary if the recommendations are accepted?
- How can the UPR process influence AI regulation across regions (N-S) with different capacities? What capacity support do States need to integrate UPR AI assessments and recommendations into their policies?
Expected Outcomes
After the event, we will produce a session report with best practices for integrating UPR into effective AI governance advocacy. We will adapt this information into different formats to improve visibility across pertinent partners, networks and social media, and for futher discussion in the CADE’s Digital Governance Working Group, enabling replication in partners’ contexts. We aim for:
- Leveraging the UPR mechanism building on its proven track record to advance human-rights accountability through multi-stakeholder engagement.
- Reach consensus on strategies/resources to identify and measure AI human-rights impacts.
- Share best practices to translate the AI harms into actionable UPR recommendations.
- Provide lessons learned on UPR advocacy in North and South contexts.
Format
Roundtable
Duration (minutes): 90
Format description: A round table displaces the expert-to-audience hierarchy that silences implementation experience. When multistakeholder from the panel and the audience sit together, this configuration fosters the space to candidly surface why UPR-AI recommendations fail domestically: capacity gaps, conflicting institutional mandates, budget constraints. This peer-to-peer vulnerability is where real solutions emerge. Ninety minutes allow time for initial framing without rushing participants into dialogue, space for substantive breakout-room discussion across in-person and virtual cohorts, and a closing window to distill emergent insights without truncating the synthesis. Hybrid participation rhythms also demand buffer; this duration prevents the false efficiency of compressing complexity, allowing genuine deliberation on how to make the most of the UPR for AI accountability.
Duration (minutes): 90
Format description: A round table displaces the expert-to-audience hierarchy that silences implementation experience. When multistakeholder from the panel and the audience sit together, this configuration fosters the space to candidly surface why UPR-AI recommendations fail domestically: capacity gaps, conflicting institutional mandates, budget constraints. This peer-to-peer vulnerability is where real solutions emerge. Ninety minutes allow time for initial framing without rushing participants into dialogue, space for substantive breakout-room discussion across in-person and virtual cohorts, and a closing window to distill emergent insights without truncating the synthesis. Hybrid participation rhythms also demand buffer; this duration prevents the false efficiency of compressing complexity, allowing genuine deliberation on how to make the most of the UPR for AI accountability.
Hybrid Format: Collective Brainstorming: On-site participants contribute via post-its; online participants via Miro board. A designated support person will transfer inputs between spaces. This exercise will nourish the session's conversations with inputs.
Breakout Rooms: To encourage candid participation, we will create parallel breakout spaces: at least one cohort for participation at the venue, and one remote breakout on Zoom. Depending on the number of participants, we will divide them into additional groups. Each group explores the same policy question through their regional and institutional lens, then reconvenes to share distinct insights. On-site and remote speakers will support their respective cohorts.
Open Floor. Now that participants have broken up the ice, we will open the floor to revisit the central ideas we want to capture, so they are well established for the reporting and follow-up steps. A designated support person will ensure remote participants who want to comment are given proper space.
Organizer 1: Angela Alarcon, Forus
Organizer 2: Iwanska Karolina, European Center for Not-for-Profit Law
Organizer 2: Iwanska Karolina, European Center for Not-for-Profit Law
Speaker 1: Patricia Ainembabazi , Intergovernmental Organization, African Group
Speaker 2: Abed Kataya, Civil Society, Asia-Pacific Group
Speaker 3: Line Gamrath Rasmussen, Government, Western European and Others Group (WEOG)
Speaker 4: Gina Romero, Intergovernmental Organization, Intergovernmental Organization
Speaker 2: Abed Kataya, Civil Society, Asia-Pacific Group
Speaker 3: Line Gamrath Rasmussen, Government, Western European and Others Group (WEOG)
Speaker 4: Gina Romero, Intergovernmental Organization, Intergovernmental Organization
Moderator
Angela Alarcon, Civil Society, Latin American and Caribbean Group (GRULAC)
Online Moderator
Iwanska Karolina, Civil Society, Western European and Others Group (WEOG)
Rapporteur
Angela Alarcon, Civil Society, Latin American and Caribbean Group (GRULAC)
What will participants gain from attending this session?
