B1: Most Important Skill in the Age of AI

The Explanation

Last week, Maria, who works at a small shipping company in Manila, had a busy afternoon. A storm had delayed several trucks. Customers were calling every few minutes. Her manager asked her to update a large customer in Singapore. The customer was important. If they felt ignored, they could change suppliers.

Maria asked an AI tool to write a short update email. She typed, “Write a polite message. Say the delivery will arrive on Thursday. ” The AI wrote a smooth email in seconds. It sounded calm and professional. Maria almost clicked “send. ” Then she looked at the schedule again. The truck was not arriving Thursday. It was arriving Friday morning.

If Maria had sent the email, the customer might have planned workers and storage for the wrong day. That could cost money and trust. Maria fixed the date and added one clear detail. She wrote, “Friday, 9 a. m. , because the storm closed one highway. ” The email took one extra minute. That minute protected her company’s reputation.

Stories like Maria’s are common now. In many offices, people use AI for first drafts. They use it for emails, meeting notes, and short reports. The news is not that AI exists. The news is how companies are changing their training. They are teaching people to write better instructions and to check the results. Managers are learning that speed without judgment can create new risks.

One skill that teams talk about is “leverage. ” Leverage means you get more results with the same effort. Maria can write three emails in the time she used to write one. But leverage has a price. If the instruction is vague, the AI may guess. It may choose a wrong date, a wrong name, or a wrong promise. The email can still sound confident.

That is why prompt quality matters. A prompt is your instruction to the AI. A weak prompt is short and unclear. A strong prompt has a goal, a reader, and limits. Maria now uses prompts like this: “Write a 90-word update for a logistics manager. Use a calm tone. Include the exact delivery time: Friday 9 a. m. Do not invent reasons. ” The AI cannot read her mind. She must give it a target.

Maria also learned to break big tasks into smaller steps. One day, her manager wanted a short report about late deliveries. At first, Maria asked the AI for the full report. The answer was too general. Now she uses a simple workflow. First she pastes real data from the tracking sheet. Then she asks for three sections: the main problem, the cause, and the next steps. After that, she asks the AI to rewrite each section in clearer English.

She also uses a small “loop” to improve the text. She asks, she reviews, and she asks again. The first version is only a draft. The second version becomes clearer. The third version becomes shorter. Maria might say, “Make this more friendly,” or “Make it less emotional,” depending on the customer. This saves time, but it still needs her eyes and her choices.

The most important habit is verification. Maria checks facts before she sends anything. She checks dates, names, prices, and promises. When AI writes, it can sound sure even when it is wrong. Maria has one rule: if the message contains a number, she checks it in the company system. If it contains a reason, she checks it with the operations team. AI helps her write. It does not give the final truth.

Practice