This weekly update explores how governments and organizations are creating new rules to control AI agents. These are computer programs that can make their own decisions and take actions without asking people first.

Singapore has taken a big step by launching a new Model AI Governance Framework for Agentic AI. This framework helps companies understand how to use AI agents safely and responsibly. The framework addresses special challenges that come with AI agents, like making sure these systems don't make mistakes that could harm people or businesses.

The European Union continues to enforce strict rules for AI systems. The EU AI Act requires companies to test their AI systems carefully and watch them closely. This is especially important for AI used in important areas like jobs, money, education, and government services.

There is a big difference in how countries handle AI regulation. The United States is focusing on letting companies innovate quickly, while Europe wants stricter safety checks. This creates different rules in different parts of the world.

Experts say that current governance frameworks need improvement to handle AI agents properly. The frameworks that work for regular AI systems don't work as well for AI agents that make their own choices. Organizations need clearer rules about what decisions AI agents can make on their own and who is responsible when something goes wrong.

Companies are now using policy as code, which means writing rules in computer language that AI agents must follow. This helps make sure AI agents stay within boundaries and don't do things they shouldn't do. These systems create records of everything the AI does, which helps companies check if AI agents are following the rules.