Artificial Intelligence & Machine Learning
,
Next-Generation Technologies & Secure Development
The Non-Binding Resolution Calls for ‘Safe, Secure, and Trustworthy AI Systems’
The United Nations on Thursday unanimously adopted a U.S.-brokered first-ever international resolution promoting the safe use of artificial intelligence.
See Also: Freeing Public Security and Networking Talent to do more with Automation
More than 120 countries, including China, co-sponsored the non-binding resolution, which calls for “safe, secure, and trustworthy AI systems” and seeks to hold signatory nations accountable for any misuse of AI systems that could potentially undermine human rights.
Thursday’s vote marked the end of a three-month negotiation led by the United States Ambassador to the UN Linda Thomas-Greenfield.
“The resolution we just adopted reflected our shared responsibility and collective fate. It leaves no one behind. It lays the groundwork for AI systems that leave no one behind, either,” Thomas-Greenfield told reporters after the vote.
The adoption of the resolution comes at a time when governments, including the United States and China, are vying to exert influence and lead global governance in the nascent technology.
The European Union has attempted to capture the regulatory space through passage of the AI Act. The United Kingdom held a first global conference on AI governance, and U.S. President Joe Biden signed an executive order on AI that directs developers of advanced artificial intelligence models to notify the government and share the results of safety tests (see: White House Issues Sweeping Executive Order to Secure AI).
“This consensus revolution, developed with direct inputs from more than 120 countries and co-sponsored by more than 120 member states from every region, is a landmark effort and a first-of-its kind global approach to the government and the use of this powerful emerging technology,” U.S. Secretary of State Anthony Blinken said.
In addition to encouraging states to adopt safe AI practices, the resolution calls for multi-stakeholder involvement and recognizes that malicious design in AI applications could lead to biases, undermine information integrity and pose a threat to human rights.