Quiz collegato alla guida · Medio Livello

Quiz sull'intelligenza artificiale per i bibliotecari

Scopri come le biblioteche possono utilizzare l'intelligenza artificiale per la scoperta, i metadati e il lavoro di consultazione, proteggendo al contempo la privacy, citando le fonti e mantenendo i bibliotecari disponibili.

Domanda 1 di 8

What should a library discovery chatbot provide with a factual answer?

Anteprima di ogni domanda8 domande

Le risposte e le spiegazioni vengono visualizzate mentre rispondi al quiz.

  1. What should a library discovery chatbot provide with a factual answer?
    • Links to source records and a path to librarian help
    • A confidence score as a replacement for sources
    • Only a generated paragraph
    • A claim that it searched every available work
  2. Why can a generated citation be risky?
    • A citation in a familiar style guarantees that it supports the answer
    • A real book title proves the system verified the cited passage
    • It may be inaccurate or refer to a source that does not exist
    • An author name in a citation confirms that the source exists
  3. What privacy concern applies to library search queries?
    • Queries are always anonymous by default
    • They can reveal sensitive interests or personal questions
    • Search history cannot be stored by a tool
    • Only book titles can be sensitive
  4. How should a chatbot answer a question outside its trusted scope?
    • Invent a likely local policy
    • Use the newest answer from social media
    • Hide which sources it searched
    • Escalate to a librarian or clearly state the limitation
  5. Which error can arise when AI generates library catalog metadata?
    • It may misidentify people or erase relevant cultural context
    • Metadata cannot be corrected by library staff
    • It always decreases catalog searchability
    • It guarantees consistent classification
  6. Which evaluation helps assess a discovery tool?
    • Count generated responses without checking them
    • Measure only answer length
    • Test accuracy, coverage, and escalation on representative queries
    • Use one staff query as a complete test
  7. How should a discovery assistant characterize search results?
    • Replace source evaluation and librarian guidance
    • Suggest candidate sources for users to inspect
    • Determine the truth of every retrieved source
    • Guarantee comprehensive coverage of all knowledge
  8. Why retain human reference support?
    • Human librarians cannot use search tools
    • Chatbots always return outdated information
    • Users should not ask follow-up questions
    • Some questions require context and judgment beyond automated retrieval