Sam Altman calls for a US-led forum to set global AI safety rules

In short: OpenAI chief executive Sam Altman says countries should create a US-led international forum to set AI safety standards, and elected leaders should make the rules.

What happened

Sam Altman, the co-founder and chief executive of OpenAI, wrote that the world needs a shared approach to safety for advanced AI systems. He said he recently discussed this at a G7 meeting with leaders from major democracies and people from the AI industry.

Altman proposed a US-led international forum that would set accepted safety standards, test advanced AI systems, and publish impartial expert analysis about risks. He said countries could join by agreeing to follow the rules, and companies inside those countries would go through regular certification, like passing safety inspections.

He also suggested the forum could help keep AI from becoming too concentrated in the hands of a few companies. He compared the idea to international cooperation in areas like aviation safety, global financial standards, and the International Atomic Energy Agency, which oversees nuclear technology.

Altman warned that if global safety standards are not created, more countries may choose restrictions on AI instead. He also argued that AI labs should not be the ones making the most important decisions. In his view, citizens and their elected representatives should set the rules, not a small number of companies.

Why it matters

Advanced AI is already being used by governments and big companies, and Altman expects even more powerful systems within a year or two. If countries cannot agree on shared safety checks, people may see uneven rules, limited access, or higher risks from misuse. A common set of standards would be like having the same traffic rules in many places, so people know what is allowed and what is not.

Source: Financial Times