Bài kiểm tra có liên kết hướng dẫn · Trung bình Cấp độ
Khám phá lời nhắc và nhật ký trò chuyện của AI dưới dạng bài kiểm tra bằng chứng
Kiểm tra khả năng phát hiện, đặc quyền, bảo quản theo tỷ lệ so với sản xuất, trạng thái sau này của một đơn hàng OpenAI và xử lý nhật ký AI có trách nhiệm.
Câu hỏi 1 của 8
A litigation hold covers a team’s use of an AI drafting tool. What should the organization identify first?
Xem trước mọi câu hỏi8 câu hỏi
Câu trả lời và giải thích xuất hiện khi bạn làm bài kiểm tra.
- A litigation hold covers a team’s use of an AI drafting tool. What should the organization identify first?
- Only final documents filed in court, never drafts or source records
- Relevant prompts, outputs, settings, users, retention systems, and related records
- Every employee’s unrelated personal AI use, regardless of relevance
- Only model weights, because prompts are not electronically stored information
- What did the October 9, 2025 stipulation change about the May OpenAI log-preservation order?
- It ended the general going-forward preservation duty effective September 26, while previously segregated data remained preserved subject to exceptions
- It ordered all AI providers to produce every user conversation to the plaintiffs
- It required OpenAI to delete every previously preserved log immediately
- It made the May order a permanent nationwide retention rule
- Does an attorney’s use of an AI tool automatically make the prompt privileged?
- No, because work product cannot include electronic records
- Yes, if the prompt contains a legal word such as “analysis”
- No; protection depends on context, purpose, confidentiality, and applicable law
- Yes, any prompt typed by a lawyer is privileged
- Which limits apply when a party requests prompts and chat logs in federal civil discovery?
- Relevance and proportionality alone; privileged material is discoverable if electronically stored
- Relevance, proportionality, privilege, and applicable case-specific orders
- The provider’s ordinary retention setting, without a litigation-specific review
- The device used to enter the prompt, regardless of relevance
- Why record the model or tool version and relevant settings for an AI output used as evidence?
- To prove the output cannot be challenged by another party
- To help explain and authenticate the system context that produced the output
- To guarantee the output is factually correct
- To establish that the prompt is privileged
- Which statement best distinguishes preservation from production?
- A preservation hold automatically waives all privileges in the records
- Production must include every preserved item even if irrelevant or privileged
- Preservation is required only after a court orders a specific chat export
- Preserving potentially relevant records does not itself decide whether they must be produced
- A request seeks logs for every user over several years. Which response best applies proportionality?
- Assess relevant users, time period, claims, burden, privacy, and targeted search or sampling options
- Collect all users’ logs and produce them without review because they are searchable
- Delete older records before discussing the request with opposing counsel
- Search only for the term “AI” and treat that result as a complete response
- If a litigation hold covers relevant AI logs that ordinary settings would soon delete, what should the organization do?
- Follow ordinary deletion because product retention settings override a hold
- Preserve every unrelated user's records without considering the hold's scope
- Preserve relevant records as required by the current hold while minimizing unrelated data
- Keep only model weights and delete covered prompts and outputs