To Appreciate the Power of AI Agents, Build a Morning Brief

Companies have spent tens of billions putting AI in front of their people. The returns are not showing up. This year MIT studied 300 enterprise AI efforts. About 95 percent had produced no measurable impact on the bottom line. Only 5 percent were creating real value.

The surprising part is that the models are good. The tools that failed were aimed at the wrong work, or at work no one had clearly defined. Most never fit how people actually work. Access went up. Value did not.

I see the same thing one level down, with people. Someone gets a license and opens the chat box. They ask it to clean up an email and decide AI is fine but oversold. No one showed them where to start. So they poke at random tasks, feel let down, and go back to how they worked before.

I tell people to build a morning brief. It is the fastest way I know to feel what an AI agent can do. It takes the part of your morning you should not be doing and hands it back to you.

The usual advice is to just experiment. Play with it, and value will follow. For most people it does not. Playing with no target gives you a pile of party tricks and no habit.

The other common answer is a big training push. A prompt library, a lunch-and-learn, a deck of examples. That teaches people about AI. It rarely changes what they do on a Tuesday.

What changes behavior is picking one real task and handing it off well. Not the flashiest task. The right one. Most people have never been told how to choose it.

The way I choose is a simple test. I have been building it into a larger model I call shift and lift.

Shift the repeatable work to an agent. The reading, the gathering, the first draft, the watching for changes. Lift yourself onto the work only a person can do. The judgment, the conversation, the decision, the relationship.

The part people miss is that the payoff is not the task the machine finished. It is what you do with the time you get back. Saved time is not the same as value. Left alone, it fills back up with more of the same. You have to spend it on purpose, on work that moves an outcome.

The test has three parts. One, AI is already as good as an average person at the task, or better. Two, the task eats real time. Three, the time you free can go straight into something that matters.

Most people miss the third part. They automate something, feel clever for a minute, and pour the saved time back into their inbox. Shift without lift is just a faster treadmill.

Most of us start the day the same way. You open your laptop and start piecing the day together. What came in overnight. What is on the calendar. Which deals moved. What you promised someone on Thursday. What is on fire now.

This is real work, and it is slow. McKinsey estimates that people spend close to two hours a day just searching for and pulling together information. That is almost a quarter of the workday. For most of us, the first slice of it happens before we have done anything that counts.

Run that morning against the test. Is an agent as good as you at scanning and sorting? Yes, and faster. Does it eat real time? Every day. Can the freed time go somewhere that matters? Yes. Straight into the first real conversation of your day.

That is a clean one to hand off. So I did.

The first thing I do most mornings is read a brief an agent wrote for me while I slept.

It reads my email, my calendar, my docs, and my meeting notes. Then it hands me a short page. The one thing that matters most today. What is overdue. What is waiting on me. What moved since yesterday. For my sales work, it flags the deals at risk and the follow-ups I owe.

It took me about half an hour to set up. That is not a project. That is a coffee break.

What changed was not the few minutes I saved. It is that I stopped starting the day underwater. I used to open my inbox and let it set my agenda. Now I start with a clear read of what matters. I walk into the first meeting already thinking about the meeting, not digging for context on the way there.

That is the lift. The brief is not the win. The win is the better conversation I have because I got there with a clear head. The agent did the sensing. I kept the judgment.

You might think this is just a to-do list with extra steps. It is not. A to-do list holds what you already decided. The brief finds what you have not seen yet. It looks across every place your day is hiding and tells you what matters first.

A morning brief is not one fixed thing. It bends to your role. The point stays the same for everyone. Hand off the assembly. Keep the decisions.

If you run your own work, keep it simple. The most important task today. What is overdue. What people are waiting on you for.

If you lead a sales team, aim it at the pipeline. Deals slipping. Where things went quiet. Who to follow up with. A quick prep for today's calls.

If you run marketing, point it at the funnel. How campaigns are doing. What a competitor shipped. Where the numbers moved, and what is stuck waiting on you.

If you manage people, add one layer. What your team needs from you today. Where you are the bottleneck.

Same habit, different reading. Everyone ends up in the same place. You start the day already lifted into the work that counts.

You do not need a technical setup or a new tool. About thirty minutes with whatever AI you already have. Claude, ChatGPT, or Copilot all work.

The shape is simple. Connect what it can see: your email, calendar, docs, and pipeline if you sell. Paste in a prompt that says what a brief means to you. Give it your role and what you care about this quarter. Then run it every morning for a week and correct it. It gets sharper as it learns you.

To make it painless, I put the exact prompt and a version for each role into a short guide. You open it, follow the instructions, and the AI does the rest. Want it? Reply and I will send it over. No catch. I would rather more people feel this than not.

Most people are still waiting to be impressed by AI from a distance. You get it by handing off one real job and watching your morning change. Start small. The brief takes you off the worst part of your day, and it does it every day. Once you feel that, the next thing worth handing off is easier to spot. That is what shift and lift is about, and what I will write about next. Handing off the right work so your time goes where it actually pays.

Your first brief will not be perfect. That is fine, and expected. Read it. Cut what you do not need. Tell it what it missed. Keep shaping it until it feels like yours. Getting real value from AI is a back and forth, and that skill is the part most people never build.

Build yours this week.

Sources: MIT NANDA, "The GenAI Divide: State of AI in Business 2025." McKinsey Global Institute on time spent searching for information.