“For developing economies, artificial intelligence opens up the possibility of overcoming outdated technologies and directly serving the people who need it most. But the potential for positive transformation it offers remains difficult to grasp and, without considering possible catastrophic scenarios, it is already clear that malicious use of artificial intelligence (AI) can undermine trust in institutions, weaken social cohesion and threaten democracy.” UN Secretary-General António Guterres argued for the need to create an advisory body on AI last year: High Level Advisory Body on Artificial Intelligence (HLAB-AI) or High-Level Advisory Body on Artificial Intelligence. This entity, made up of 39 prominent AI leaders from 33 countries, has published this Thursday the seven fundamental measures (Final report) to ensure inclusive AI development that, in Guterres’ words, “benefits all of humanity” and avoids risks.
In the final report of the HLAB-AI, released today and titled Governing AI for Humanitymore than 2,000 people have participated who have already made a preliminary work to call for the establishment of a cooperative and inclusive global AI architecture to “foster the development and protection of all human rights.”
The advisory body regrets the absence of “a global framework governing AI” and the lack of commitment from member states; only seven are part of all the initiatives (Canada, France, Germany, Italy, Japan, the United Kingdom and the United States) and “118 are missing entirely, mainly in the Global South,” according to the final text.
“With its development in the hands of a few multinational companies in a few countries, the impacts of AI implementation risk being imposed on the majority of people without them having a say in the decisions to do so,” warns the HLAB-AI in its report.
To avoid this, the organization proposes seven measures that it considers “agile, adaptable and effective”:
International scientific panelThe aim of this body will be to “provide unbiased and reliable scientific expertise on AI”. Its mission will be to help Member States develop a shared vision of the new tool and address information asymmetries between laboratories and the rest of the world. The panel will produce annual reports on AI capabilities, risks and trends, along with specific reports on emerging risks.
Political dialogueThis measure will convene intergovernmental and multi-stakeholder meetings, including technology companies, users and national entities, to foster regulatory common ground that guarantees human rights.
International AIThe body proposes connecting representatives of regulatory organizations, technology companies and civil society to ensure the technical interoperability of AI systems across borders.
Common benefitsA global network will promote access to the developments of researchers and social entrepreneurs through training, resources and data.
Global Fund. It will be responsible for addressing capacity and collaboration gaps.
Global Data Framework. It aims to standardize definitions, principles and data management, ensuring transparency and accountability in AI systems.
AI AgencyThe report proposes the creation of a small AI office within the United Nations Secretariat to support and coordinate the implementation of these proposals.
Some entities in the business sector agree with the UN criteria. A study by the Institute for Business Value IBM (NYSE:IBM) believes that “there is an urgent need for a new type of technological leadership in the era of generative AI.” A survey of 2,500 international executives reveals that the most advanced tools will determine the greatest competitive advantage, according to 72% of those surveyed. Also for the majority (80%), “transparency in their organization’s use of next-generation technologies, such as generative AI, is essential to foster trust.”
Hildegunn Kyvik Nordås, a professor at Örebro University in Sweden and a researcher at the Norwegian Institute of International Affairs, also agrees with the UN’s approach. “The astonishing capabilities of AI applications have raised deep concerns about the future of work, democracy and even humanity itself, as well as a sense of urgency to establish effective governance at all levels and use cases. It has also inspired visions of a bright future in which prosperity, good health and a clean environment are the norm,” she warns in a publication of the UN. Economic Policy Council (CEP) after the work prior to the final report.
Given this scenario, the member of the CEP, a non-partisan, non-profit organization, warns about developments towards “applications capable of self-learning and performing tasks beyond what they were initially trained for,” skills known as General Artificial Intelligence. Kyvik Nordås admits that they are “probably decades away and many doubt whether it will ever be developed,” but the researcher believes that “it makes sense for special UN agencies to develop global governance systems within their fields.”
The organization Partnership on AI (PAI) is also a non-profit association made up of academic, civil society, industry and media organizations that create solutions for AI to promote positive outcomes for people and society. Its researcher Jiyoo Chang shares the sense of the work of the HLAB-AI, but warns: “The rapid advance of the development and adoption of artificial intelligence urgently demands the translation of principles into practical guidance. This requires collective action and coordination.”
