The United Nations advisory body focused on AI governance released its new report Governing AI For Humanity this week. Co-chaired by Carme Artigas, a Spanish official who previously served as secretary of state for Digitalisation and AI, the panel assesses the likely global trajectory of artificial intelligence and flags a key risk: power could concentrate in the hands of a few players.
The long term impact of AI can affect many people. If governance is left to a small group of developers or to the countries that host them, a deeply unfair situation would arise where the consequences of creating, deploying and using AI would fall on the majority without a voice in decisions.
The report highlights the enormous positive potential of these technologies. They could open new avenues for scientific research, optimize energy networks, improve public health and agriculture, and accelerate progress toward the Sustainable Development Goals. Yet governance must be distributed and actively monitored to prevent the concentration of power.
If regulation is lacking, AI opportunities may fail to materialize or be distributed unevenly. The growing digital divide could confine benefits to a small set of states, companies and individuals.
To address these issues the plan calls for creating a small AI office within the United Nations Secretariat. This office would support and coordinate the implementation of proposals and act as a unifying mechanism to align diverse initiatives across regions and sectors.
The UN advisory body is the most representative assembly of 39 experts worldwide able to reflect humanity’s AI aspirations. It is co-chaired by Carme Artigas and James Manyika, senior vice president at Google Alphabet. The report outlines a strategy to mitigate AI risks and urges the United Nations to establish the foundation of a first governance framework for this technology that is inclusive, globally distributed and rooted in international cooperation.
Up to 118 countries outside AI governance
Furthermore, the document outlines a series of recommendations to narrow current gaps in AI governance and calls on governments and all stakeholders to unite around governance efforts to promote responsible development, deployment and protection of human rights worldwide.
At present, seven of the 193 United Nations member states participate in the leading AI initiatives, while 118 nations, largely in the Global South, remain outside due to the absence of a comprehensive global framework.
There are differences among countries and sectors, but a strong desire for dialogue persists. Involving diverse experts, lawmakers, business leaders, researchers and advocates from different regions and disciplines shows that diversity does not have to block agreement and that open conversation can yield common ground and collaboration.
[–>
The panel recommends establishing a shared understanding and common benefits, supervised by the small UN AI Office, which would help coordinate the implementation of these proposals and act as a cohesive force to link initiatives in a practical, sustainable way.