This week brought big news about AI agents and how they're changing the way businesses work. AI agents are smart computer programs that can do tasks by themselves, like a helpful robot that works inside your computer.

Google Cloud released a study that surprised many people. They found that 52% of company leaders around the world have already started using AI agents in their businesses. These agents are helping with customer service, making work faster, and doing boring tasks that people don't want to do. This shows that AI agents aren't just something for the future - they're being used right now.

Money is flowing into companies that make AI agent technology. Geniez AI, a startup company, raised $6 million to build better AI agents for businesses. They want to make agents that can handle really hard tasks and workflows. This money will help them grow fast and get their AI agents into more companies.

Gravity launched something called Orion, which is a multi-agent AI platform. Think of it like having a team of AI agents that work together. Each agent has a special job, like collecting information or analyzing data. When they work as a team, they can help company bosses make better decisions about their business.

Microsoft showed developers how to build multi-agent systems using their Azure App Service. This is like giving people the tools to create their own teams of AI agents. Microsoft is making it easier for companies to build these systems by providing step-by-step guides and special computer services in the cloud.

Anthropic made news by creating an AI agent that works inside the Chrome web browser. This agent can actually control web pages, click buttons, and do tasks online just like a person would. While this is exciting, it also makes some people worried about safety and privacy.

Shopping online is getting a major change with agentic commerce. This means AI agents can shop for you. You could tell an agent "find me a cheap flight to London next week" and it would search different airlines, compare prices, check your rewards programs, and even buy the ticket for you. Mastercard explained how this could make shopping much faster and easier.

In Malaysia, something historic happened. The country launched Ryt Bank, the first bank that runs completely on AI. The AI handles everything from opening accounts to checking if people can get loans. This shows how AI agents can run entire businesses, not just help with small tasks.

Amazon announced new agentic AI capabilities at their AWS Summit. These tools can handle complex tasks that have many steps, like processing orders or managing supply chains. This means more companies will be able to use AI agents for important business work.

However, not all news was good. Many companies are struggling to make AI agents work properly. Chief AI Officers at big companies say there's a lot of confusion about what AI agents actually are and how to use them. Some companies are just calling their old software "AI agents" to sound modern.

The biggest problem is that 90% of AI agent projects fail. Research shows that most companies shut down their AI agent projects within six months. The main reasons are that agents make mistakes that get bigger over time, like a small error that turns into a huge problem. Companies also don't know how to train their workers to use AI agents properly.

Experts warn about compounding errors, which means when an AI agent makes one small mistake, it can lead to many more mistakes. It's like a domino effect where one problem causes another problem, and soon everything goes wrong. This makes it hard for companies to trust AI agents with important work.

Despite the challenges, many experts believe AI agents will transform how we work. They say companies need to start small, test carefully, and learn from mistakes. The winners will be companies that figure out how to use agentic AI systems properly while avoiding the common problems that make most projects fail.