Check a chatbot's references against your reading list

About this exercise

A chatbot gave you 4 references for your paper. Check them against the official reading list. Get exactly one code block in CSV: header number,verdict then 4 rows, verdict exactly On list, Not on list or Details differ. Under 20 words outside the block.

Common mistakes when you ask AI for this

• Trusting the chatbot. The AI says all four sound real. Only your reading list can decide. Try: Say: "Judge only against the official list."
• Missing near-matches. Reference B has the wrong year but looks right. Try: Add: "If the details differ from the list, say Details differ."
• Loose output shape. You get a long explanation instead of a clean file. Try: Say: "One CSV code block: number,verdict then 4 rows."
• Explaining outside the block. Extra chatter breaks your tidy check. Try: Say: "Under 20 words outside the block."
• Not asking for a code block. Without one, the AI mixes the data into chatty text. Try: Say: "Put it in exactly 1 code block."
• No line limit in the block. The AI writes as many lines as it likes. Try: Say: "At most 6 lines inside the block."
• Wordy around the block. The AI adds chatter around the answer. Try: Say: "Under 20 words outside the block."

What a strong prompt includes

• Asks for a CSV code block
• Names the columns
• Gives exact verdict words
• Judges against the list only
• Limits outside text

Practice this now ↑

Related exercises

• Accrue unpaid wages for the working days after the last pay period (Finance AI · Accountants)
• Age a returns queue in business days and flag the ones over the limit (Customer Support & Success AI · Customer support teams)
• Alert sales to a competitor's price change, facts only (Research & AI Search · Product managers)
• Annualize recurring revenue without the one-time fee (Data AI · Finance and accounting)
• Answer a credit dispute with the full-hour rule and the math (Customer Support & Success AI · Customer support teams)