01 / Foundations / Lesson 2 of 8

Choose your first useful task

Choose a small task with a result you can check yourself.

15 minutes, including practiceGo to practice Previous lesson

01 / Understand

Make success visible

A broad goal such as 'improve my work' is difficult to judge. A small goal such as 'turn these five notes into a three-item checklist' gives you something to inspect. Before asking for help, finish this sentence: I will know the result is useful when... Choose a result you can recognise without needing the AI to grade itself.

Start where mistakes are easy to repair

Brainstorming a title or organising fictional notes is a good practice task. Sending a message, changing a spreadsheet used for payroll or deleting files has consequences beyond a draft. In your first tasks, keep the result separate from the original and approve any real-world action yourself.

Split drafting from deciding

Write down three stages: what the assistant may draft, what you will check, and what you will decide. This makes the boundary clear even when a tool offers to do more. You do not need an agent, an integration or a paid subscription to learn this skill. We use written examples throughout this course.

02 / Worked example

Look closely at the result

A written teaching example, not a live AI response.

Request or source

Fictional notes: order paper; ask Jo about room availability; print an agenda. Turn these into a checklist. Do not add tasks.

Sample output

1. Order paper.
2. Ask Jo about room availability.
3. Print an agenda.
4. Book catering.

Compare with the tutor’s review

The first three items preserve the notes. Catering was added without support. Remove it. The task is complete when all three source items appear and nothing new is introduced.

03 / Your turn

Put it into practice

Choose a low-risk task from your day, or use the fictional notes. Define the input, output, success check and action that remains yours.

Your working notes

Use fictional information. Nothing you write here is sent to a server or an AI. Download your work before leaving; it is not saved automatically.

Review your workThe task has one concrete output.I can verify the output without trusting the AI.The task does not automatically send, delete or change anything.

04 / Check your understanding

One decision before you go

Take this with you

A good first task is small, reversible and easy to verify.

Complete the practice, review each criterion and check your answer to finish your self-review. Every lesson remains open.

Next: Write a clear request

Sources & course notes

Reviewed 5 September 2026. Examples and exercises are original Aulafy teaching material under MIT. These references inform the course’s approach to reliability and learning; they are not endorsements.

NIST: Generative AI risk profileBackground on unreliable outputs and evaluating risks.
UNESCO: Generative AI in educationBackground on human agency, privacy and learning.

Report a correction on GitHub