This weekly update covers big changes in how artificial intelligence agents will be controlled and used around the world. On March 20, the White House released a new policy framework for AI that recommends Congress create one set of rules for the entire United States instead of letting each state make its own rules. The framework has seven main goals, including protecting children and creators, helping American companies be the best at AI in the world, and making sure AI is not used to hide information from people. The Trump administration believes having one national rule is better than many different state rules. However, experts say Congress might not actually make this framework into law right away. Meanwhile, in Europe, the EU AI Act continues to get stronger, with rules that require companies to keep humans in charge of important AI decisions. Singapore also released its own guidelines for controlling AI agents, which many organizations around the world are using as examples. These different approaches by different countries show that AI regulation is becoming a global challenge where different parts of the world are taking different approaches. Companies that work in multiple countries will need to follow different rules in each place. The main focus for most regulators is making sure AI agents—the kind of AI that can make decisions and take actions on its own—stay safe and follow rules.