US Rallies Support at UN to Align ‘Safe’ AI Systems Globally

Augusta Saraiva and Antonia Mufarech
·3 min read

(Bloomberg) -- The United Nations General Assembly approved a resolution to promote “safe, secure and trustworthy” artificial intelligence systems, one of the most high-profile efforts yet to establish global standards for the rapidly rising disruptive technology.

Most Read from Bloomberg

The non-binding proposal, which was introduced by the US and co-sponsored by more than 110 countries including China and India, encourages members to support “responsible and inclusive” AI development through domestic regulations and governance. It passed by consensus on Thursday without a formal vote.

US President Joe Biden highlighted the need to regulate AI in a speech before the General Assembly in New York in September, and was a top agenda item when he met China’s leader Xi Jinping in November. Still, Washington’s push at the UN contrasts with a lack of strong regulations at home despite calls for action by lawmakers and businesses, including Open AI. China has implemented strict guidelines, while the European Union has approved the most comprehensive list of AI rules globally.

Read More: Regulate AI? Here’s What That Might Mean in the US

Washington had tasked dozens of embassies with lobbying governments worldwide, according to a senior US official who asked not to be identified discussing private matters. Countries including Russia and Iran have engaged in the negotiations, another official said.

“The resolution squarely addresses the priorities of many developing countries, such as encouraging AI capacity building and harnessing the technology to advance sustainable development,” US National Security Advisor Jake Sullivan said in a statement, adding that negotiations took nearly four months.

The resolution doesn’t cover the military use of AI, another growing concern for the international community as powers like the US and China rush to incorporate it into their armed forces. Washington has used AI to identify targets in the Middle East this month, and Israel is also relying on new technologies in its war against Hamas.

Read More: Artificial Intelligence Is Already Changing How the US Wages War

“It is our hope that this resolution does not represent the end of our collaboration around this critical issue,” said Linda Thomas-Greenfield, the US ambassador to the UN. “We hope the inclusive and constructive dialogue that led to this resolution will serve as a model for future conversations on AI challenges in other arenas, for example with respect to peace and security and responsible military use of AI.”

The UN’s AI advisory board established last year by Secretary-General Antonio Guterres warned in a recent report that governance remains “territorial and fragmented,” and called on countries to step up domestic and international regulation efforts to avoid “accountability deficits.”

Meantime, Volker Turk, the UN human rights chief, has also been ringing the alarm over dangers posed by the non-regulated use of AI. During a trip to California last month, Turk engaged in multiple debates on the issue, and held meetings with influential figures in the AI space including OpenAI Chief Executive Officer Sam Altman.

Among other guidelines, the new UN resolution calls for the protection of privacy and personal data through the development of “effective safeguards,” including physical security and risk management systems. It also focuses on closing the gap between richer and poorer countries, calling on UN members to “urgently” engage with developing nations on issues such as technology transfers, technical assistance and financing.

The resolution also highlights the protections for human rights in the context of AI systems, and “reaffirms that the same rights that people have offline must also be protected online.”

(Updates with comment from fifth paragraph.)

Most Read from Bloomberg Businessweek

©2024 Bloomberg L.P.