Campus History — Part 1: AI Draft + Critique

This is Part 1 of a two-part project, althought the second part is much more involved. For part 1, you generate an AI history of a UNM topic and write a short critique / analysis of it.

Skills you are practicing

Source criticism — evaluating a source by asking: who made it, for whom, in what context, and what does it leave out? You’ll do this every time you encounter a historical claim in a news article, a political speech, or an AI summary.
Reading for absence — noticing not just what a source says but what it doesn’t say, and asking why. A confident-sounding account with no admission of gaps or uncertainty is a serious red flag.

AI on this assignment

The AI history (Step 2): built in. Use any AI tool, and paste its response exactly as it came out.
Your critique (Step 4): no AI. Don’t use it to write, polish, or brainstorm the critique—including pasting the AI history into a second chatbot and asking what’s wrong with it. Spotting that is the assignment.

A hard project, on purpose

The Campus History project asks you to do a lot of different things at once: pick a topic, critique an AI source, dig through boxes in an archive, learn enough GitHub and Markdown to publish a web page, choose and caption images, and turn all of it into a story someone would want to read. None of these pieces is out of reach on its own. What makes the project hard is juggling them, and keeping going when one of them stalls—the archive doesn’t have much, the site won’t update, the story won’t come together.

That difficulty is one of the real values of the project. Most work worth doing outside of school looks like this: several unfamiliar parts, no single right answer, and a stretch in the middle where you’re not sure it’s going to work. Learning how to get through that is a skill in itself:

Break it into pieces. Part 1 and Part 2 already do some of this for you. Keep going: one step at a time, in order.
Start early. Most of the frustration comes from discovering a problem the night before it’s due.
Ask for help sooner than feels necessary. Librarians, classmates, and I can usually solve in five minutes what can eat an evening on your own.
Expect to feel lost at some point. That’s a sign you’re learning something new, not a sign you’re failing.

And then you finish. At the end you’ll have a published essay, built on research nobody else did, that anyone can read. Succeeding at something hard—and knowing how you did it—is worth carrying into whatever you take on next.

Step 1: Pick a topic

Make sure you’ve completed the steps at Campus History student guide to make YOUR COPY of the Campus History site that you will work on and create your own essay folder.

Step 2: Generate an AI history

Use an AI tool (ChatGPT, Claude, Gemini, or similar) to generate a ~600-word history of your topic. Give it a specific prompt:

“Write a ~600-word history of [your topic] at the University of New Mexico. Include specific dates, events, people, and sources where possible. Be clear about your sources.”

Save the full AI response without editing it. This is your baseline document. You’ll use this is phase two of the Campus History project when you compare this to what you find in the archive. Save your AI essay on your webpage. Paste it in under the header metadata, after the ---.

Also note which tool you used and whether it searched the web while answering (most tools show a searching step or list links when they do). A tool that searches will cite real pages and invent less, which changes what you’ll find in Part 2. Put that note at the top of your critique: ChatGPT, model, thinking level, web search on, for example.

Step 3: Read it critically

Read the AI draft with a skeptical eye. Read carefully and ask:

What sources does it cite? To what extent do you trust them?
Where does it seem to make claims about the past that don’t seem supported by evidence?
How much does it provide larger cultural context for understanding the topic?
What seems plausible and specific? What seems vague, generic, or suspiciously confident?
What questions does it raise that it doesn’t answer?
Whose perspectives are emphasized? Whose are left out?
What makes it sound like AI writing?

Step 4: Write your critique

Write a short paragraph (~350 words; general refraction length) evaluating the AI draft. Be specific — don’t just say “it seemed accurate” or “it made things up”. Spot check information it provides against whatever sources you can find online (obviously use the ones AI cites, but you can find others, too).

This critique is short, but it’s significant for the assignment. It’s asking you to do what historians always do: read a source critically before deciding how much to trust it.

Obviously, don’t use AI to write the critique. There should be a distinct diffrence in writing styles between the AI-essay and your analysis.

Post this to the discussion board like a refraction post.

What I’m looking for in your critique

This is a refraction-like assignment graded on engagement and honesty, not on whether the AI draft turns out to be accurate. What makes a strong critique:

Most importantly, address the questions in step 3. You’ll have more to say about some questions than others, and you may answer some at the same time. You don’t need to go one by one, but your reponse as a whole should address those questions.
You quote specific lines from the AI draft — not just “it seemed vague” or “it invented sources”
You report one claim you actually checked, and what happened when you did
Your thinking is genuine even if uncertain: “I’m not sure if this is accurate but here’s why it feels suspicious…” is an excellent observation
You notice the vagueness: could this paragraph apply to almost any other topic if you swapped out a few words?

There is no wrong answer here. What matters is the quality of your attention to the AI response.

A — quotes specific lines; distinguishes between claims that are checkable and ones that can’t be verified without specialist knowledge; reports one claim you actually checked; reasoning is genuine even when uncertain
B — specific examples; some analytical depth beyond surface impressions
C — addresses the prompt but stays general; examples are thin or paraphrased rather than quoted
D — vague summary with no textual evidence from the AI draft

Submit

Your AI-generated draft should be on your own campus-history page.
Your ~350-word critique, posted to the appropriate Canvas discussion topic, like refractions are.