Responsible AI Line
Bias, accuracy, transparency, copyright and security in practical terms, with checklists you can apply this week.
All levels4–5 hours5 stationsFree
Who it's for: Everyone who publishes, decides or advises with AI help — especially in regulated or customer-facing work.
What you'll be able to do
- Recognise where bias enters an AI-assisted decision
- Disclose AI use appropriately
- Understand the basics of AI and copyright
- Reduce prompt-injection and data-leak risk
The route
Station 1: Bias in, bias out
How training data and prompts shape outputs, with a simple fairness spot-check.
Station 2: Accuracy and accountability
Who is responsible when AI is wrong, and how to document human review.
Station 3: Disclosure and trust
When and how to tell customers, readers or students that AI helped.
Station 4: Copyright basics
What is generally understood about AI outputs and source material, and when to ask a lawyer.
Station 5: Security hygiene
Prompt injection, data leakage and safe handling of AI-connected tools.
Read along
- AI hallucinations: a five-question habit for catching errors · 8 min
- How to write a one-page AI use policy (with template) · 8 min
- Using AI at work without leaking sensitive data · 8 min
Watch along
Deep Dive into LLMs like ChatGPT
Andrej Karpathy · Foundations
A long, general-audience walkthrough of the full training stack behind chat assistants.
Intro to Large Language Models
Andrej Karpathy · Foundations
A one-hour talk on what LLMs are, how they're trained, and where they're heading.
How to complete this line
Work through one station per sitting. For each: read the linked guide, watch the video where one is listed, then apply the idea to one real task from your own work and note the result. When you have done all stations, try the matching brief in this month's prompt challenge.