F-100 · Adult · Foundations
Foundations of AI
8 workshops · ~12 hours
What a language model actually is, why it can be wrong, and how to use it without becoming dependent on it. Eight workshops, no math required, no programming required.
The workshops
What you do, step by step
01
What a language model actually is
Strip away the marketing. See what the model is and isn't.
Artifact: A one-page mental model you can explain to a colleague.~90 min
02
Why models hallucinate
The mechanism behind confident wrong answers.
Artifact: A prompt-and-response notebook where you reproduce three hallucination patterns.~75 min
03
Prompting that earns its weight
Six prompt patterns that consistently improve output, and one anti-pattern to abandon.
Artifact: A reusable prompt library for your daily work.~90 min
04
Reading the output critically
How to spot the difference between confident-and-correct vs confident-and-wrong.
Artifact: A checklist you can apply before you trust any LLM output.~60 min
05
Context windows and memory
Why the same model behaves differently at different points in a conversation.
Artifact: A diagram of how context flows through a multi-turn exchange.~75 min
06
Tool use, in plain English
How models can take actions in the world — and why that's different from text generation.
Artifact: A worked example of an agent calling a real tool.~90 min
07
Where AI breaks
The seven failure modes you'll encounter in production use of any model.
Artifact: A failure-mode catalog for your team's first AI deployment.~75 min
08
What comes next
An honest read on the next 18 months — capability gains, safety questions, what to keep watching.
Artifact: A short essay you write articulating your own AI-literacy plan.~90 min
Subscribe
One subscription, every track.
$39/month or $390/year (Standard) covers every SOSI track in the catalog — today and as new ones ship. Family is $69/month or $690/year and covers three seats.