Learn AI / Learning Paths
Learning paths
Each path is six to ten terms in a deliberate order for one situation. Mark a term as understood on its page and it fills in here. Progress is stored in this browser only, so there is nothing to sign in to and nothing to lose track of.
AI basics in 15 minutes
The nine words everything else is built from. Start here if you have never started anywhere.
Understand ChatGPT and large language models
What is actually happening when a chat assistant answers you, and why fluent is not the same as correct.
AI terminology for diplomats
The vocabulary that shows up in negotiated text, and what each word commits anyone to.
AI for managers and front offices
Enough to chair the meeting, ask the right question, and know what must not be pasted into a chat box.
AI for programme and policy officers
How a prediction is made, how it is judged, and how it quietly goes wrong once it is in the field.
AI for human rights practitioners
Where the harm actually lands: in the data, in the threshold, and in who never gets to appeal.
AI for communications teams
Generated media, provenance, and the difference between a model being wrong and a person lying.
AI for data and digital teams
The build-side vocabulary: what you index, what you retrieve, and what you measure afterwards.
Understand AI governance
Who writes the rules, what kind of rules they are, and why "not binding" is not the same as "no effect".
Understand AI risks without panic
The failures that actually happen, in proportion, without either hype or doom.
How AI learns
From a pile of examples to a working model, one mechanism at a time. The engine-room tour.
From prompts to agents
How a system that answers becomes a system that acts, and what has to be supervised when it does.
From embeddings to RAG
Why searching by meaning works, and how it turns a confident guess into a cited answer.
How to assess an AI proposal
The ten questions to ask a vendor, a colleague, or a concept note, in the order you should ask them.
AI terms appearing in UN resolutions
Words that made it into negotiated text. Each one links to the milestone where it landed.