Abrusci

A successful Dialogue would agree on some basic commitment in terms of relevant frameworks to apply and a designation of the relevant bodies that will be entrusted with the enforcement and oversight to move the conversation further.

Protection and promotion of human rights;Transparency, accountability, and human oversight

Human rights are continuously restricted as a consequence of AI application. Although it is been widely recognised that the same human rights recognised offline must be equally respected and protected in the digital sphere, there is a persistent gap between the promises of these high-level legal principles and provisions and the lived experiences of users. It follows that it is of utmost importance to not only reiterate that existing human rights framework applies, but provides concrete avenues for enforcement and application, which include adequate oversight, transparency requirements and accountability mechanisms. Existing scholarship shows that without these concrete procedural safeguards, any regulatory effort remain ineffective.

The identification of the institutions that would contribute to AI governance is missing. It is important to concretely discuss the potential role of multi-level governance models applied to AI governance, leveraging existing national, regional and international institutions.

The recent regulatory frameworks adopted in Europe (such as the EU AI Act, the CoE Framework Convention on AI, etc) add significant protection to fundamental rights in terms of high-level commitments and somehow pave the way to appropriate enforcement. However, this needs to be embedded in a global governance framework to avoid jurisdictional issues to serve as an excuse for lack of compliance. Coordination among international, regional and national institutions is key for effective enforcement and governance and the UN should take the lead in moving this forward with concrete proposals and pushing on the economic incentives for a human-rights based approach to AI governance

The AI Dialogue should support the development of a multi-level governance approach to AI. This requires: 1) reviewing existing AI governance initiatives at the national and regional level; 2) identify best practices, lessons learned and obstacles to further action; 3)establish a dialogue between private actors and public bodies to understand needs and challenges in ensuring appropriate governance and enforcement; 4)partner with academia to develop models for multi-level governance. This is a project I'm currently working on and I would be happy to assist if needed.

The AI Dialogue should build on existing regulatory opportunities as presented by the UNESCO Guidelines on Platform Governance, the CoE Framework Convention on AI and HR, the EU AI Act, the EU DSA and works that National Human Rights Institutions across the world have done (see, for instance the toolkit prepared by the African NHRIs and ICNL.

The Dialogue should include thematic workshop and general agora panels where all stakeholders are invited both in person and online.