Responsible AI Use In Practice – Module 5 Implementing AI Safely Action Plan
📍 Module 5: Implementing AI Safely – Action Plan
đź“ť Overview
This module offers practical ways to document and review how AI is used in your business or team — without needing complex systems. It also introduces the idea of identifying risk levels, keeping simple records, and knowing when to get advice.
Whether you’re just using a handful of tools or exploring more advanced applications, the goal here is to help you feel organised and in control.
🎯 Learning Objective
By the end of this module, you’ll be able to:
-
Identify where AI tools are being used in your work or team
-
Understand what questions to ask about those tools
-
Create simple records or checklists to support transparency
-
Recognise higher-risk uses that may need expert support
-
Know how to plan for future regulatory and organisational changes
đź“– Text version of the lesson
Why have an action plan?
Responsible AI use isn’t just about rules — it’s about awareness. By keeping track of how AI is used and how decisions are made, you’re helping reduce risk, increase trust, and improve future decision-making.
This doesn’t need to be complicated — small steps can have a big impact.
Four helpful starting steps
-
Identify where AI is being used — even built-in tools like Gmail suggestions or Microsoft Copilot.
-
Understand what each tool does and what kind of data it uses.
-
Check if human judgement is still being used to sense-check results.
-
Record your use cases — a short list or table is often enough.
Questions to help assess risk
-
Is this tool supporting or replacing a human decision?
-
Could the output affect someone’s rights, money, access to services, or wellbeing?
-
Are you confident in how the data is being handled?
-
Does the use feel proportionate, fair, and explainable?
These aren’t legal tests — just good prompts for awareness.
Simple checklist or log structure (example)
You could create a short document that lists:
-
Tool name and provider
-
Purpose or business use
-
Data type involved
-
Whether outputs are reviewed
-
Any known concerns or notes for follow-up
This can help you build a lightweight, internal “AI register” — useful if someone new joins the team or a regulator ever asks questions.
When to ask for help
You don’t need to manage this alone. It’s a good idea to speak to a Data Protection Officer, IT lead, or specialist if:
-
Your AI tool is used in recruitment, finance, education, or legal advice
-
You’re unsure what data it’s using or where it’s going
-
The tool seems to influence decisions about people or access to services
Support available locally
-
Digital Isle of Man can offer signposting and guidance
-
Vendors often publish technical and risk documentation
-
Specialist advisers (legal, data protection, ethics) can help for more complex or public-facing uses
-
Your internal governance team may already be tracking digital tools and risks
Looking ahead: preparing for change
The EU AI Act and other regulations are being phased in over the next two years.
Starting with a simple plan now helps you:
-
Identify potential gaps
-
Build shared understanding
-
Adjust gradually rather than react quickly later
🪞 Reflective Prompt (Optional)
Think about the AI tools you’ve personally used in the last few weeks.
Question:
Do you know:
What those tools are doing with your input?
Whether anyone else is reviewing or logging their use?
Who you’d speak to if you had a concern?
If not, that’s okay — this is the perfect time to start building a clearer picture.
âś… Suggested Next Step
We recommend setting up a shared note, spreadsheet, or table with your team listing:
-
What AI tools are in use
-
What each is used for
-
Whether it involves personal data or business decisions
-
Whether it’s reviewed by a person
This basic register can support transparency and help you align with upcoming legal and ethical expectations — without requiring technical expertise.