Aulafy02 / Everyday practice / Lesson 5 of 8
Protect your information
Prepare a practice input without exposing personal or confidential details.
01 / Understand
Use only the information a task needs
A request to improve the tone of an email does not need a customer's real name, phone number or account details. For learning, invent the whole example. Removing a name alone may not be enough: a rare job title, address or unusual event can still identify someone.
Aulafy and external tools are different
The exercise below runs in this page. It does not send your writing to an AI model. If you separately paste text into another app, that service's policies apply. Do not assume a free plan, a private-looking chat or a delete button means information is never retained. Check the service and your organisation's rules before using real work material.
Keep permissions small
Reading a draft is different from giving software access to your inbox or files. Before connecting a tool, identify what it may read, change and send. Practise without integrations first. A local model may reduce some cloud transfers, but the app can still use network features or save files; 'local' is not a complete privacy guarantee.
02 / Worked example
Look closely at the result
A written teaching example, not a live AI response.
Fictional exercise: Please rewrite a customer complaint containing a full name, home address and order history. The task is only to make the tone polite.
Use an invented customer and a generic delayed delivery. Leave out the address and order history. Ask for a neutral draft without making a refund promise.
Compare with the tutor’s review
The replacement preserves the writing task while removing details it does not need. For this exercise, do not paste the original complaint at all.
03 / Your turn
Put it into practice
Create a fully fictional complaint in two sentences and a request to rewrite it politely. List three kinds of information you deliberately left out.
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.
04 / Check your understanding
One decision before you go
Take this with you
For practice, invent the data. For real work, check policy and minimise access.
Complete the practice, review each criterion and check your answer to finish your self-review. Every lesson remains open.
Next: Improve a draft without losing factsSources & course notes
Reviewed . 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.