The UN General Assembly on Thursday adopted the first resolution on artificial intelligence, which is aimed at ensuring the powerful new technology benefits all nations, respects human rights and is “safe, secure and trustworthy”.
The resolution, co-sponsored by 123 states, was adopted by consensus and without a vote, meaning it had the support of all 193 UN member nations.
The resolution “promotes international co-operation to manage the risks of AI, because we know that unless AI is developed responsibly and used responsibly, it can lead to accidents and it can compound threats from malicious actors”, senior US administration officials told reporters in a briefing before the resolution's passage.
The resolution, which is the result of three months of intensive negotiations, involved the consideration of thousands of suggested edits and feedback from more than 120 countries, the officials added.
When questioned about whether there was opposition to the resolution from Russia or China, which also endorsed it, the US officials acknowledged there were “lots of heated conversations” but said Washington “actively engaged with” Beijing and Moscow.
After the adoption of the resolution, the US ambassador to the UN, Linda Thomas-Greenfield, told reporters in New York that she has been asked over the past few months why the UN should involve itself in AI regulation.
“The answer is straightforward,” Ms Thomas-Greenfield said.
This AI tool can help you speak any language – video
This AI tool can help you speak any language
“AI is already being used to detect disease and predict natural disasters. It is helping farmers grow more food and educators reach more students … it is helping activists and members of civil society protect our democracies,” she said.
Morocco’s ambassador to the UN, Omar Hilale, said AI offers numerous potential benefits for developing countries in various fields of sustainable development.
“That is why we have joined forces with the United States from the first day on this important process, with the aim to enhance a constructive dialogue on AI among EU and member states and to help ensure that this fast-evolving technology serves the common good of humanity,” he said.
The European Parliament last week adopted the Artificial Intelligence Act, which is considered the world's first comprehensive horizontal legal framework for the technology.
The regulation is expected to come into force at the end of the legislature in May, following the completion of final reviews and approval from the European Council.
The US has been pressing politicians to adopt AI regulation measures, but a polarised Congress has made little headway.
“We now look forward to building on this landmark achievement,” the US National Security Adviser Jake Sullivan said in a statement following the passing of the resolution.
“As this technology swiftly evolves, we will continue to strengthen international co-operation and respond to the far-reaching implications of AI.”
