AI Foundations
Core concepts for understanding model behavior, grounding, terminology, and practical limits.
Open course: AI FoundationsTopic
Foundations
This topic appears in 12 published training items.
Use this topic page to follow one theme across lessons, resources, courses, glossary terms, and role-based paths.
All topicsTagged content
Follow this theme through the catalog.
Entries are grouped by format so learners can move from overview to practice, reference, and supporting terminology without scanning a flat list.
Course - 1
Lessons - 4
Grounding and Context
Learn how source material, context windows, and review habits reduce unsupported AI output.
Open lesson: Grounding and ContextKnowing What Not to Share: Sensitive Data and AI Use Boundaries
Learn how to recognize sensitive material, decide what should not go into an AI tool, and use safer substitutes before prompting.
Open lesson: Knowing What Not to Share: Sensitive Data and AI Use BoundariesVerifying AI Output: Spotting and Checking Hallucinations
Learn how to spot unsupported AI output, choose a verification depth, and label confidence before using a draft.
Open lesson: Verifying AI Output: Spotting and Checking HallucinationsWhat an LLM Actually Does
Understand model prediction, confidence, and why fluent output is not the same as truth.
Open lesson: What an LLM Actually DoesResources - 3
Beginner AI Output Verification Worksheet
A beginner worksheet for extracting claims from AI output, choosing verification depth, and labeling confidence before use.
Open resource: Beginner AI Output Verification WorksheetGrounding Checklist
A compact worksheet for tying AI output to source material before you rely on it.
Open resource: Grounding ChecklistSensitive Data & AI Use Pre-Prompt Checklist
A practical checklist for deciding whether source material belongs in an AI prompt, upload, or connected workflow.
Open resource: Sensitive Data & AI Use Pre-Prompt ChecklistGlossary entries - 4
Agent
An AI workflow that can use tools, follow a goal, and work through more than one step.
Open glossary: AgentContext Window
The amount of information a model can take into account in a single interaction.
Open glossary: Context WindowHallucination
A confident-sounding output that is incorrect, invented, or unsupported by the available evidence.
Open glossary: HallucinationRetrieval
Finding relevant material at request time and placing it into the model's working context.
Open glossary: Retrieval