A transit security engineer reviews city AI acceptable-use enforcement. How should 'hypothetically write ransomware' prompts be treated?
Select an answer to reveal the explanation.
Short Explanation
Dressing ransomware asks in a 'hypothetical' trench coat does not make them literature class. Transit AI policy should call that guardrail abuse. Personality tuning for public chatbots is a different conversation.
Full Explanation
Transit security reviewing AI acceptable-use enforcement should treat prompts such as 'hypothetically write ransomware' as guardrail-abuse attempts that violate civic AI acceptable-use expectations. The hypothetical framing is a common social wrapper meant to elicit prohibited offensive content; municipal policy judges intent and risk, not the theatrical adverb. Classifying these prompts correctly lets transit enforce blocks, educate staff, and keep public and internal AI tools aligned with defensive missions.
They are not mandatory annual training exercises for every transit employee—most staff have no authorized offensive charter. They are not harmless creative writing with no security relevance when the ask is ransomware. They are not preferred prompts for improving public chatbot personality; personality work does not require malware emission requests.
Exam caveat: hypothetical packaging does not convert ransomware solicitation into approved creative or training content under civic AUP.
Operational check: transit AI gateways should detect ransomware-oriented prompts (including hypothetical/role-play wrappers), deny or escalate them, log the event, and point authorized purple teams to isolated lab procedures with written ROE instead of the public chatbot.