Measuring AI output quality
Data and documentsWorking with AI
What you will be able to do
- Build an evaluation set with normal, edge and failure cases. Compare actual results with an explicit scoring rubric.
Core concept
- Preserve source meaning and distinguish data, formulas, formatting and generated interpretation. A file that opens is not automatically accurate, editable or compatible with the target application.
Practice 1 · Try a small task
- Create a synthetic table with A: 12, B: 8 and a missing value. Produce a small artifact suited to this lesson, preserving the source and marking the unknown. Compute the known total independently: 20.
Practice 2 · Failure and verification
- Change one value and check recalculation, layout or extraction in the target application. Include an empty row and a long label. For documents, inspect rendered pages; for spreadsheets, compare formulas and recalculated values.
Review your work
- Keep the source, editable output, independent expected result and visible verification. Identify any information that was inferred rather than present in the input.
Use synthetic or anonymized material for practice. A person must review before real sending, payment or deletion.
These guide exercises do not award completion, XP or badges. Complete the existing verified labs and quiz in the full lesson to earn them.
The full lesson requires sign-in and course access.
Full lesson in Korean →