This weekly update covers major changes in how countries are making rules for AI agents - special computer programs that can make decisions and act on their own. These AI agents are becoming more powerful and companies need to understand new laws.

Ireland is creating a new law to handle AI agents with a special government office called the AI Office of Ireland. This office will watch over AI systems and make sure they follow the rules. In the United States, two states - California and Texas - started new AI laws on January 1, 2026. These laws ask companies to be honest about how their AI agents work.

South Korea also made one of the first major AI laws and started it in January. When AI agents make mistakes or cause problems, people are asking an important question: Who is responsible? Is it the person who built the AI? Is it the company using it? Or someone else? Courts and lawyers are working to answer this question.

Another big change is about trust and safety. Companies need to make sure AI agents explain why they make choices and what they will and will not do. Contracts between companies are also changing because AI agents work all the time, not just when someone tells them to. The rules are still being figured out, but one thing is clear: AI agents need better supervision and clearer boundaries.