Artificial Intelligence Discussed at the United Nations
World leaders and diplomats are meeting to discuss global issues, including artificial intelligence (AI).
The rapid development of AI, which started with the introduction of ChatGPT, has impressed many. However, some experts have raised concerns about the possible risks of AI, such as spreading false information or AI systems becoming uncontrollable.
The United Nations is working to create a system to manage AI. Previous attempts to control AI by countries like Britain, South Korea, and France have only resulted in voluntary promises.
The General Assembly recently agreed to create two groups focused on AI: a global forum and a panel of experts.
The UN Security Council will hold a meeting to discuss how AI can be used responsibly, following international laws, to support peace.
The UN Secretary-General will also host a meeting to start the Global Dialogue on AI Governance, a place for governments to discuss working together and sharing ideas. The forum is scheduled to meet in Geneva next year and in New York in 2027.
Efforts are underway to find experts for the scientific panel.
Some experts believe that these new groups show progress, but might not have enough power. They wonder if the UN can effectively manage AI, since the technology is changing so quickly.
Ahead of the meeting, some experts are calling for governments to agree on clear rules for AI by the end of next year. They believe these rules are needed to prevent the most dangerous risks.
This group, which includes people from companies like OpenAI and Google, wants governments to create an international agreement on AI. They point out that the world has previously agreed to treaties banning nuclear testing and biological weapons.
One expert, Stuart Russell, suggests that AI governance could be similar to how the International Civil Aviation Organization works, coordinating with safety regulators in different countries to ensure they follow the same standards.
Instead of creating strict rules, diplomats could create a flexible framework that can be updated as AI continues to advance.
