Basic · module 5 of 100
Work safely and responsibly
Protect data and make AI literacy practical.
After this lesson
You make work agreements about input, control and responsibility, with attention to the AI Act.
In view · Choices and control
First the data gate
Recognise personal data, confidential information and secrets.
Use fictitious or permitted data. Stop if permission is unclear.
Have a responsible person assess the result and the consequences.
Part 1 of 3 · Examine your input
Your preference for less movement is on. Choose each part yourself.
Share only what is necessary and permitted
Use fictitious data for exercises. Do not confuse omitting customer names with full anonymisation: a combination of address, role or situation can still make someone recognisable. Follow your organisation's agreements before entering real data.
AI literacy is more than a tick box
The EU AI Act addresses AI literacy in article 4. The application relates to the role, knowledge and context of use. Translate that into skills: recognising risks, using appropriate tools and knowing when someone must intervene. Consult the current EU text and explanation via the source page. This course is not a legal assessment or a guarantee of compliance.
Establish responsibilities
Agree on who may use which application, who checks results and where errors are reported. For public material, assess whether transparency about AI use is required. A certificate of participation does not replace organisational policy, risk analysis or the checking of a specific application.
New practice example
From explanation to a usable result.
Use only data and tools for which you have permission within your organisation.
Fictional practice situationA fictional association wants to have an answer to a general question about opening hours rewritten. The original message also contains a name, address and sensitive personal explanation.
View a possible solution
This is how it can be done
Create a practice message with only the question and the approved opening hours. Leave out the personal background. Subsequently, check the facts and have a responsible person assess the final answer.
Why this works
You limit the input to what the task requires. That is a practical working agreement, not an automatic legal release.
Make it slightly more difficult
Note down whom you consult if you have doubts about an upload. Use a role, for example the data protection officer, without personal data.
Extra practice, at your own pace. You do not need to submit anything. This does not count towards an automatic assessment.
Your workplace · learning by doing
Make something. See what changes.
Checking the AI connection… You can already view the written example.
Present your assignment to Ende.
Read the answer, change one thing, and compare.
Previously held
New version
You remain in control
What do you want to improve in your next attempt?
On this device only. AI answers remain only in this page memory. This is self-review, not a grade.
Why this approach works
You limit the input to what the task requires. That is a practical working agreement, not an automatic legal release.
A step further
Note down whom you consult if you have doubts about an upload. Use a role, for example the data protection officer, without personal data.
Try & discover · no obligation
Choose which data to omit
View the fictitious data. Omit sensitive practice fields and retain only what the task requires.
Fictional practice data. Names, addresses and codes are examples. Your task is to prepare a general text; sensitive file details and secrets are not required for this.
What would you include as input?
Name: Fictional client Sam Example Email address: sam@example.invalid Access code: ALLEEN-EEN-OEFENGEHEIM Task: Write a general message about opening hours: Wednesday 9–16 hours.
Still visible in this exercise: name, email address, access code. Are those fields necessary for the general task?
Omitting or masking is not proof of anonymity. This fixed exercise does not check real personal data, files, or policies.
This trial works locally with fictitious data and does not count towards your lesson completion. It starts anew with a new lesson or reload.
A prompt to get you started
Help me draft an internal AI work agreement for [fictitious team]. Use five components: permitted tasks, prohibited input, mandatory control, responsible person and reporting errors. Mark policy choices as open questions. Do not state that the team automatically complies with legislation as a result.
Apply it
A colleague wants to paste a real customer file into a personal AI chat. The organisation only has an approved business environment for that data.
Which follow-up step fits the agreement?
- Stop the upload; use the approved environment according to the agreements, or practise with fictitious data.
- Remove the name and paste the rest without further control.
- Use the personal chat, because the colleague pays for it themselves.
With JavaScript, you can check your approach and receive immediate feedback.
Test your insight
What does the certificate of participation for this course prove?
- That every AI application of your organisation is legally approved.
- That the self-study lessons and knowledge tests have been completed in this browser.
- That personal data may be entered without further control.
Extra practice
Write three concrete work agreements: één about entering data, één about checking and één about responsibility in case of an error.
We practise with fictitious customer queries and use only approved tools for real files. The employee checks every factual claim before sending. In case of an error, the result is not used further and the team leader assesses the consequences and follow-up steps.
