Question 3: CCAR-P: Claude Certified Architect - Professional
A technical team is cataloging risks specific to Claude’s use in a document-grounded Q&A system. Which two items represent failure modes that are intrinsic to LLM-based systems rather than generic software defects? (Choose two.)
Nội dung câu hỏi
A technical team is cataloging risks specific to Claude’s use in a document-grounded Q&A system. Which two items represent failure modes that are intrinsic to LLM-based systems rather than generic software defects? (Choose two.)
Các lựa chọn
Đáp án được giữ gọn theo nhãn A, B, C, D trong phần bình chọn tương tác.
- A. the model generating a plausible-sounding answer that is unsupported by any retrieved document — đáp án hiện tại
- B. the model refusing a legitimate query because surface features triggered an overly broad safety pattern — đáp án hiện tại
- C. an expired TLS certificate blocking outbound API calls to the Claude endpoint
- D. a database connection timeout causing a retrieval call to return an empty result set
- E. a misconfigured load balancer routing requests to a deprecated API version
Answer researched independently against supporting documentation; not a published source answer key. Independent reasoning (not a published answer key): Unsupported plausible answers and inappropriate safety refusals are model-behavior failures; TLS and database faults are infrastructure failures. References: https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-hallucinations https://www.anthropic.com/research/demystifying-evals-for-ai-agents
Cộng đồng
0 bình luận công khai. Tên thành viên được ẩn một phần.