Working artifact · built for Squirrel Space

Knowing the words is not the same as knowing what to ask.

I built a two-level workplace AI Literacy Diagnostic for Squirrel Space. It separates familiarity with AI vocabulary and tools from the ability to understand, question, verify, and participate in workplace AI decisions.

Take 101 →Take 201 →

The diagnostic originated through Squirrel Space. The version running here is the current unofficial CFG concept adaptation. No account needed. No email collected.

101 · Working Understanding

Can you follow the conversation?

Common vocabulary: prompts, models, agents, and what each word actually means
How models behave, and why a confident explanation can still be wrong
Hallucination, citations, and what a cited source proves by itself
Context and token limits when you dump everything into one request
Writing a prompt whose result can be checked
Confidentiality: what is safe to paste into a public AI tool
Verification before relying on an answer about a spec or a number
APIs and retrieval at a working level
Prompt injection and why pasted content is not neutral
15 questions, immediate feedback after every answer

Start this level →

201 · Applied Understanding

Can you challenge the conversation?

Tokens, context windows, and what long inputs cost in practice
Model settings such as randomness, and what each one trades away
Chat UI versus API, structured outputs, and tool calls
Retrieval and RAG: similarity that is not an answer, and citations that do not support the claim
Fine-tuning versus retrieval when the underlying data keeps changing
Model memory and why relying on it is risky
Source systems and how their state limits the AI built on top
Evals and why a good demo is not production readiness
15 questions, immediate feedback after every answer

Start this level →

What this is

A practical diagnostic.

It tells a person, a team, or a manager where the gaps are before those gaps show up in a deployment decision. It is useful the way a pre-flight check is useful.

What this is not

Not a credential.

No certification, no scientific validation claim, no score to put on a resume. It is an assessment built to be honest about what someone can currently reason about.

How this fits the current work →Talk about using it with a team