Library › AI Evals › Product managers
File a bug report for an AI translation that broke a price
Practice: Asking for a format you can use. About 5 minutes.
About this exercise
Your AI translated a German product page and turned the price into the wrong number. Write exactly 6 numbered lines: Title, Input (the source text), Expected, Actual, Severity and Likely cause. Use the localization team's scale: Sev1 = a price, legal or safety text is wrong, Sev2 = meaning changed, Sev3 = wording only. Base the cause on the logs, not guesses. Plain text, under 160 words.
Common mistakes when you ask AI for this
- Guessing the severity. The AI picks 'Critical' or 'High' on its own scale, not the localization team's. Try: Give the scale: "Sev1 = a price, legal or safety text is wrong."
- Fuzzy expected vs actual. The AI writes 'price was incorrect' without the two numbers. Try: Say: "Expected and Actual give the exact prices and the job id."
- Repeating the Slack guess. The AI echoes 'the AI is bad at German' instead of reading the log warning. Try: Add: "Base the cause on the logs, not Slack opinions."
- A long incident write-up. The AI adds impact, timeline and next steps until it's a page long. Try: Be exact: "Exactly 6 numbered lines."
What a strong prompt includes
- Lists the fields
- Gives the severity scale
- Points to the logs
- Says exactly how many
- Asks for plain text
Related exercises
- Announce a referral bonus with the steps to claim it (Marketing AI · HR and recruiting)
- Announce a team meeting change people will actually follow (Writing AI · Managers and team leads)
- Answer a 'found an error' question in STAR order (Writing AI · Finance and accounting)
- Answer a wholesale inquiry with clear terms (Writing AI · Small business owners)
- Answer an RFP timeline question in exactly four phases (Writing AI · Sales and customer success)