Purpose: Summative evidence for CO-01 (analyse trade-offs) and CO-04 (creative direction with tools). Judgement: Satisfactory (S) / Not Satisfactory (NS) only—no percentage grade.
Time guide: 45–60 minutes. Submit in your provider’s format (document, LMS text box, or PDF).
Assessment tasks
Task 1 — Scenario: AI-assisted vs hand-authored (CO-01)
Scenario: A colleague wants a three-page static site (home, about, contact) for a side business. They can spend one weekend on it and may use any AI tool their workplace allows.
In 200–300 words, analyse trade-offs of AI-assisted versus mostly hand-authored HTML/CSS for this situation. Address at least three of: speed, quality, maintainability, ethics (e.g. accuracy, disclosure, data policy). State a clear recommendation for this colleague and why, in plain language.
Task 2 — Creative direction vs mechanical typing (CO-04)
In 4–6 sentences, contrast creative direction (goals, audience, constraints) with mechanical typing of code. Give one example of a creative-direction decision that is not the same as “writing more lines of HTML.”
Task 3 — When fundamentals still matter (CO-01)
List three distinct situations where basic HTML/CSS literacy still changes outcomes—even when AI drafts the first version. One sentence each (no paragraph required).
Marking guide (Satisfactory / Not Satisfactory)
| Criterion | Satisfactory (S) | Not Satisfactory (NS) |
|---|---|---|
| Task 1 — Trade-off analysis | Discusses ≥3 relevant factors; recommendation is consistent with the analysis; writing is understandable at ~Year 10 level. | Fewer than three factors, or no clear recommendation, or mostly generic claims without application to the scenario. |
| Task 1 — AQF depth | Shows judgement (weighs pros/cons), not only bullet lists of facts. | Only lists tool names or repeats course slogans without analysis. |
| Task 2 — Contrast | Clearly separates direction from typing; example is specific and plausible. | Confuses the two ideas, or example is missing or irrelevant. |
| Task 3 — Situations | Three distinct, plausible situations (e.g. accessibility check, fixing structure, better prompts). | Fewer than three, or duplicates, or vague (“when things go wrong”). |
Overall module judgement: S only if all rows are S at an acceptable Diploma-level standard. NS if any row is NS.
What good looks like (examples)
These are illustrative — write in your own words. Do not copy verbatim.
Task 1 (extract):
For a weekend side-business site, AI-assisted work wins on speed for first drafts — I could have three linked pages live by Sunday afternoon. The trade-off is verification time: I would still check heading order, contact details, and workplace AI policy before publish. Hand-coding would teach me more about maintenance but might not finish in one weekend. Recommendation: use AI inside the IDE for structure and copy drafts, then read every file before deploy — speed with oversight.
Task 2:
Creative direction is deciding what the site must achieve and for whom; mechanical typing is entering tags. Choosing “one
h1for the business name and contact in the footer on every page” is direction — it is not the same as pasting 200 lines the model returned.
Task 3 (three situations):
- Fixing skipped heading levels before publish.
- Catching a wrong phone number the model invented.
- Writing a tighter prompt because you know what
<nav>should contain.
Exemplar responses (for assessors)
Task 1 (shape only): Look for named trade-offs (e.g. speed vs verification time), mention of hallucination or policy, and a reasoned pick—not “AI always wins” or “never use AI” without nuance.
Task 2: Accept definitions in learner’s words; example might be “choose one h1 and section order” vs “paste whatever the model outputs.”
Task 3: Accept e.g. heading order for screen readers, contrast/CSS review, debugging broken tags, writing precise prompts.