General Assembly adopts landmark resolution on AI
21 March 2024 Human rights
This is the first time that the Assembly has adopted a resolution to regulate this emerging field. The resolution calls on States to refrain from using artificial intelligence systems that cannot function in accordance with international human rights standards or put them at risk.
The UN General Assembly adopted by acclamation on Thursday a landmark resolution on promoting “safe and trustworthy” artificial intelligence (AI) systems that also benefit sustainable development for all.
By adopting without a vote a draft resolution led by the United States, the Assembly also stressed respect, protection and promotion of human rights in the design, development, deployment and use of AI.
The text was “co-sponsored” or supported by more than 120 member states.
The General Assembly also recognized the potential of AI systems to accelerate and enable progress towards achieving the 17 Sustainable Development Goals.
It is the first time the Assembly has adopted a resolution to regulate this emerging field. The US National Security Adviser stated earlier this month that the adoption would represent a “historic step” for the safe use of AI.
Same rights, online and offline
The Assembly called on all Member States and stakeholders “to refrain from using artificial intelligence systems that cannot operate in accordance with international human rights standards or that pose undue risks to the enjoyment of human rights.”
“The same rights that people have offline must also be protected online, including throughout the lifecycle of AI systems,” it said. The Assembly also urged all States, the private sector, civil society, research organizations and the media to develop and support regulatory and governance approaches and frameworks related to the safe and trustworthy use of AI.
Equity and digital literacy
The Assembly further recognized the “different levels” of technological development between and within countries, and that developing nations face unique challenges in keeping up with the rapid pace of innovation.
It also urged Member States and stakeholders to cooperate and support developing countries so that they can benefit from inclusive and equitable access, close the digital divide and increase digital literacy.
Hope for other sectors
Before adoption, Linda Thomas-Greenfield, the US ambassador and permanent representative to the UN, introduced the draft resolution. She expressed hope that the “inclusive and constructive dialogue that has led to this resolution will serve as a model for future conversations on AI challenges in other areas, for example, with respect to peace and security and the responsible military use of AI autonomy.”
Thomas-Greenfield noted that the resolution was designed to build on work already being done by the United Nations, including the International Telecommunication Union (ITU), the United Nations Educational, Scientific and Cultural Organization (UNESCO) and the Human Rights Council.
“We intend for it to complement future UN initiatives, including negotiations for a global digital compact and the work of the Secretary-General’s high-level advisory body on artificial intelligence,” she said.
We govern AI
Thomas-Greenfield also highlighted the opportunity and responsibility of the international community to “govern this technology rather than let it govern us.”
“So let us reaffirm that AI will be created and deployed in the light of humanity and dignity, safety and security, human rights and fundamental freedoms,” she said, before adding: “Let us commit to closing the digital divide within and between nations, and to using this technology to advance shared priorities around sustainable development.”
A human rights and discrimination prevention approach
From a human rights perspective, the resolution stresses that AI systems must be people-centred, ethical and inclusive, and fully respect the promotion and protection of human rights and international law. It recognises that inappropriate use of AI can undermine human rights, reinforce structural biases and increase discrimination. It therefore urges Member States to refrain from using AI systems that cannot operate in accordance with international law or that pose undue risks to human rights.
One of the most prominent risks of AI is its potential to perpetuate and amplify discrimination. The resolution urges Member States and other stakeholders to develop regulatory frameworks that include safeguards against discrimination and bias in algorithms. This involves analysing and mitigating bias contained in data sets and combating discrimination in algorithms. It also encourages the promotion of linguistic and cultural diversity in data used to train AI systems.
The resolution highlights the importance of closing the digital divide between and within countries. It also urges Member States to cooperate with developing countries to ensure inclusive and equitable access to the benefits of digital transformation and AI systems. This includes improving the connectivity of digital infrastructure and promoting digital literacy.
Gender digital divide and human oversight of AI
The resolution recognizes that there is a gender digital divide and calls on Member States to adopt specific measures to close the gender digital divide. It also calls for “incorporating a perspective that takes into account issues of disability, gender and racial equality in policy decisions and the frameworks that guide them” when using artificial intelligence systems.
It also urges them to promote human oversight of the functioning of AI-based systems by examining automated decisions. It even promotes the existence of means of redress and accountability for those affected by automated decisions.
Finally, it encourages States to develop tools such as watermarks or labelling that allow users to distinguish the origin of authentic digital content from content generated by artificial intelligence or manipulated content.
Conclusions
The adoption of this resolution represents a significant step towards promoting AI systems that are safe, secure and trustworthy. By placing a strong emphasis on human rights and sustainable development, the resolution seeks to balance the potential benefits of AI with the associated risks, promoting responsible innovation and ensuring that no one is left behind in the digital age.