Practice telling claims from evidence in a reading

About this exercise

Your Political Science professor says you keep summarizing evidence as if it were the author's argument. Get exactly 5 numbered lines, one per marked sentence in the excerpt, each labeled exactly Claim or Evidence with a reason under 8 words, plain text. Use your professor's definitions and don't summarize the reading.

Common mistakes when you ask AI for this

• Asking for a summary. The AI summarizes the reading, which is the habit your professor wants you to break. Try: Say: "Don't summarize. Label each marked sentence."
• Not giving exact labels. Without fixed labels, the AI describes each sentence instead of classifying it. Try: Say: "Label each exactly Claim or Evidence."
• Not naming the standard. The AI may call a judgment like S5 evidence because it sounds factual. Try: Add: "Use Prof. Kwan's definitions."
• Not saying how many lines. The AI picks its own number, and it usually picks more lines than you need. Try: Be exact: "Exactly 5 numbered lines."
• Trusting the AI to count words. AI is bad at counting. Ask for a total and it often lands 20% over. Try: Cap each reason: "under 8 words."
• Forgetting where it will be read. AI formats for a screen, with bold and headings that show up as stray symbols in your notes. Try: Say: "Plain text, no bold, no headings."

What a strong prompt includes

• Gives exact labels
• Uses the definitions
• Rules out a summary
• Says how many lines
• Sets a length limit

Practice this now ↑

Related exercises

• Apply three sociological perspectives to one campus example (Study AI · College students)
• Boil a Macbeth scene down to four plot lines (Study AI · High school students)
• Boil a textbook section down to 5 steps (Study AI · High school students)
• Brainstorm replies to "too expensive" in your best rep's style (Writing AI · Sales and customer success)
• Break your own conditioning example into its four parts (Study AI · High school students)