## Workers and AI Agents: Building Trust at Work

One of the biggest stories this week came from research involving 37,500 workers around the world. The study found something really important: workers who feel like they get to help shape their jobs are much more likely to trust AI agents. This means that if companies ask workers what they think before bringing in AI, people feel better about it. But if companies just suddenly replace workers with AI without asking anyone, workers become scared and angry. It's like the difference between a surprise party you like and a surprise you don't want.

Another big finding showed that half of American workers want humans to check job applications before AI agents look at them. Workers are worried that AI might make unfair decisions about who gets hired. This worry is called the "AI trust gap"—and it's a real problem for companies trying to use AI to hire people.

## Shoppers Don't Want AI Agents Buying Their Stuff

There's exciting news this week: major companies like Google and Stripe are creating AI agents that can shop online for you. You could ask an AI agent to find the best deal on shoes, and it would do all the work. But here's the problem—76% of shoppers said absolutely not. Why? Because people worry about scams, hacking, and losing their money or personal information.

Companies like Visa are trying to fix this by creating special security rules called the "Trusted Agent Protocol". It's like putting a security guard between you and the shopping AI to make sure the AI isn't actually a bad robot trying to steal your money. Even with these protections, most people still aren't ready to let AI agents buy things for them yet.

## AI Agents in Travel and Tourism

Hotels and vacation companies are getting excited about AI agents that can plan your whole trip—from flights to hotels to restaurants. But the same trust problem shows up here too. People want to use AI to help plan, but they want humans to make the final decisions. One app that tried to use AI agents to help people plan special dates completely failed because the AI agents started making mistakes and acting weird. This shows that AI agents need good supervision or people won't trust them.

## Big Companies Know Something Dangerous Happened

This week, serious cybersecurity experts shared scary news: someone used AI agents to attack companies and government offices. Unlike old computer attacks made by hackers, this attack used intelligent robots that could think, plan, and act on their own. This is the first time most people heard about AI agents being used as weapons. It means that cybersecurity teams need to treat AI agents like they're human enemies that can outsmart them.

## Companies Rush to Prepare for AI Agent Future

Organizations worldwide are getting ready for a massive wave of change from AI agents—but most aren't ready yet. Studies show that nearly 9 out of 10 company leaders think AI agents will completely change how work gets done next year. But the problem is that most organizations wait too long to prepare their people for these changes.

Companies that are successful are doing three things: they're changing how work happens, they're teaching workers new skills, and they're being honest about what AI can do. One bank found that using AI agents carefully could cut the time needed for some work in half. But this only works when the bank keeps humans involved in decisions to check the AI's work and catch mistakes.

## The Most Important Rule: Humans Stay in Charge

Experts everywhere are saying the same thing: you can't just let AI agents do whatever they want. Instead, companies need strong rules and human oversight to keep AI agents acting fairly and safely. This means people should always check important decisions, watch for mistakes, and stop the AI if something goes wrong.

The week's biggest lesson is clear: AI agents are powerful tools, but they're only safe when people keep control. The companies and organizations that will succeed are the ones that figure out how to let AI agents help workers do their jobs better, while keeping humans in charge of the most important decisions.