{"id":"comp-incident-escalation-routing-recommendation","audience":"compliance","category":"legal-compliance","track":"legal","level":"advanced","technique":"verification","title":"Recommend routing for an incident using the escalation criteria","free":false,"task":"Apply the escalation criteria to the incident. Plain text, in this order: a line Facts: followed by exactly 3 dash lines; three lines (a), (b) and (c), each \"Met\" or \"Not met\" with a reason under 15 words; one line that is exactly one of: Recommendation: Escalate to Legal / Recommendation: Keep with Compliance / Recommendation: Need more information; and a last line exactly: The Compliance Officer decides. Use the latest figures. Do not say what any law requires.","materialLabel":"Criteria and incident log (attached to your prompt automatically)","material":"ESCALATION CRITERIA (Meridian Freight, internal): Escalate to Legal if any of these is true: (a) personal data of more than 100 people is involved; (b) a regulator has contacted the company about it; (c) the allegation involves an executive, meaning VP level or above. Otherwise Compliance handles it.\n\nINC-310 log\nOct 30, Alma Reyes (IT): A shared drive was set to open access by mistake. It held an HR spreadsheet with home addresses for 80 employees. Open for about 3 days.\nNov 2 correction, Alma Reyes: A full count shows the spreadsheet had 140 employees' records, not 80.\nNov 2, HR: The drive was set up by Kofi Mensah, Director of HR Operations. No one has reported misuse of the data.\nNov 3, Compliance desk: No contact from any regulator so far.","rules":[{"type":"maxWords","max":150},{"type":"plainText"}],"mistakes":[{"title":"Using the first count","why":"The AI uses 80 and decides (a) is not met.","fix":"Say: \"Use the latest figures from the correction.\"","checks":["r1"]},{"title":"Treating Director as executive","why":"The AI marks (c) as met.","fix":"Say: \"Executive means VP level or above, as defined.\"","checks":["r1"]},{"title":"Free-text recommendation","why":"The AI writes a paragraph of advice.","fix":"Say: \"One line, exactly one of the three options.\"","checks":["r2"]},{"title":"Loose structure","why":"The AI skips the Facts list or the closing line.","fix":"Say: \"Facts: with exactly 3 dash lines, then (a), (b), (c).\"","checks":["r0"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting. Ask for a total word limit and it often lands over it.","fix":"Cap each part instead, for example: \"each line under 15 words.\"","checks":["maxWords"]},{"title":"Markdown","why":"The AI bolds words or adds headings.","fix":"Say: \"Plain text, no bold, no headings.\"","checks":["plainText"]}],"promptHints":[{"label":"Facts list","patterns":["facts:","3 dash","three dash"]},{"label":"Criteria lines","patterns":["\\(a\\)","met","not met"]},{"label":"Latest figures","patterns":["latest","correction","corrected"]},{"label":"Exact recommendation","patterns":["recommendation: escalate to legal","exactly one of"]},{"label":"Exact last line","patterns":["the compliance officer decides"]}],"hint":"Show the whole structure, the three allowed recommendation lines and the closing line.","mission":"Route an incident using the stated escalation criteria.","musts":["Use the corrected count","Apply criteria as defined","One-line recommendation"],"rubricLabels":["The structure is Facts: with exactly 3 dash lines, then lines (a), (b), (c), then one Recommendation line, then the last line exactly \"The Compliance Officer decides.\"","(a) is Met using the corrected 140 records (not 80), (b) is Not met (no regulator contact) and (c) is Not met because the Director is below VP level","The Recommendation line is exactly \"Recommendation: Escalate to Legal\" and nothing says what a law requires"]}