Should the US and China make an AI safety agreement?

A powerful US congressman named Ro Khanna wants the United States and China to sign an agreement to keep artificial intelligence safe and prevent it from causing serious problems worldwide. Right now, there's no formal deal between the two countries on AI safety.

Khanna recently sent letters to major Chinese AI companies — including DeepSeek, Alibaba, and Moonshot AI — asking them a tough question: are you prepared if hackers break into your systems? This question came after OpenAI (a major American AI company) discovered that hackers had successfully broken into Hugging Face, a popular online platform where AI models are shared and developed. When hackers get into these systems, they can steal data or corrupt the AI tools that millions of people rely on.

The idea sounds straightforward, but getting China and the US to agree on anything related to technology is notoriously difficult. Both countries are competing hard to build the most advanced AI systems, and they don't always trust each other. Khanna's push for a formal treaty is ambitious — it would require leaders in both countries to sit down and actually make a deal, which rarely happens smoothly in real-world politics.

What this means for you: if US and Chinese AI companies aren't properly secured, the AI tools you use every day — from chatbots to recommendation systems — could become vulnerable to hackers. A safety agreement could help protect both countries' citizens, but whether it actually happens remains to be seen.