Some members of the Global Digital Justice Forum—IT for Change, APC, Global Partners Digital, Derechos Digitales, and Media Monitoring Africa—collectively prepared a submission in response to a UN survey to identify the terms of reference and modalities for the establishment and functioning of the Independent International Scientific Panel on AI and the Global Dialogue on AI Governance, to be established within the United Nations.
Our submission was based on the premise that the development, deployment, and impact of AI are deeply interwoven with existing power structures, economic models, labor dynamics, and geopolitical interests. A few key highlights from our submission are below:
(i) On the AI Scientific Panel
The mandate of the UN’s Independent International Scientific Panel on AI should focus on evidence-based research and policy recommendations. It must systematically analyze AI’s societal impacts, including human rights, social, economic, political, and environmental effects. It must also take into account regional disparities in resources and technological capabilities, as well as differing impacts—including harms—on specific groups, such as Indigenous people. The Panel should contribute to the development of a unified global public interest approach to AI risk assessment, standards, policy, and regulation
The Panel should coordinate closely with bodies and processes related to AI, internet governance, and digital transformation strategies. Furthermore, it must maintain a governance structure that upholds transparency, accountability, and independence from corporate influence, with strict conflict-of-interest policies. Industry-backed stakeholders should be excluded while allowing independent consultation with the private sector
The funding mechanism of the Panel should be constituted in such a way that it prevents contributions from influencing the Panel’s mandate, agenda, or decisions while ensuring financial support for equal participation, particularly from under-resourced regions.
(ii) On the Global Dialogue on AI Panel
The Global Dialogue on AI should take place as part of, or alongside, existing dialogues such as those convened as part of the WSIS process or other international agreements and processes, including the Internet Governance Forum (IGF) and processes dealing with labour rights, human rights and environemntal sustainability. Not only should the panel be interdisciplinary, but dialogue should take place in manner that enables inter disciplinary participation
It should ground its work in international human rights law, WSIS principles, and the Global Digital Compact to ensure AI development benefits people equitably.
(iii) On the Relationship Between the Scientific and Dialogue Panels
Both entities should be institutionally independent but coordinate strategically to ensure research-driven, evidence-based AI governance. The Scientific Panel should inform Dialogue discussions with rigorous analysis, while the Dialogue Panel provides a platform for sharing findings and fostering policy coherence.
Lastly, we emphasized that the Panel and Dialogue must not used as superficial advisory bodies to delay, dilute, or sideline necessary legal and regulatory measures for AI. Strong compliance mechanisms must be built into AI governance frameworks to ensure that guidelines and recommendations translate into enforceable regulations, transparency obligations, and accountability measures.
You can read our complete submission here.