AI w 311 i quiz o chatbotach obsługi obywatelskiej

Odpowiedzi i wyjaśnienia pojawiają się w miarę rozwiązywania quizu.

1. What is 311 in many North American cities?
   • The emergency police line
   • A tax-filing portal
   • A non-emergency number and app for service requests and information
   • A voter registration system
2. What does the Open311 standard provide?
   • A common structure for service requests that tools can connect to
   • A translation engine for city websites
   • A law requiring cities to use chatbots
   • A payment system for city fines
3. What problem was reported with New York City's MyCity chatbot in 2024?
   • It was only available in Spanish
   • It could not look up addresses
   • It charged residents for answers
   • It gave answers that contradicted law, including on workers' tips and housing discrimination
4. Why do city chatbots use retrieval-augmented generation?
   • To make answers longer
   • So answers come from official pages, schedules and codes
   • To avoid needing any human staff
   • To store residents' personal data
5. Why does accurate request classification matter?
   • It raises city tax revenue
   • It changes the resident's language setting
   • A misclassified ticket can sit in the wrong department's queue for days
   • It is required to call 911
6. What should the intent classifier do with an emergency message?
   • Redirect it to 911
   • Create a standard pothole ticket
   • Ignore it
   • Answer it from the FAQ
7. How does duplicate detection typically group new tickets?
   • By the resident's age
   • By alphabetical street name only
   • By the device used to submit
   • By service type, distance and time window
8. Which statement reflects a common misconception about city chatbots?
   • Translation quality varies across languages
   • A chatbot automatically reduces workload
   • Escalation to humans helps
   • Resolution rates are a useful metric