Bài kiểm tra có liên kết hướng dẫn · Trung bình Cấp độ

Bài kiểm tra tài liệu về lạm dụng tình dục trẻ em bằng AI

Kiểm tra các khái niệm phát hiện CSAM: băm nội dung đã biết, bộ phân loại, đánh giá của con người, kết quả trùng khớp sai, lộ trình báo cáo, quyền riêng tư và các biện pháp bảo vệ an toàn cho trẻ em.

Đường dẫn hướng dẫn liên quanAI phát hiện tài liệu lạm dụng tình dục trẻ em
Câu hỏi 1 của 8

A perceptual hash matches an upload to a vetted reference entry. What does the match most directly indicate?

Xem trước mọi câu hỏi8 câu hỏi

Câu trả lời và giải thích xuất hiện khi bạn làm bài kiểm tra.

  1. A perceptual hash matches an upload to a vetted reference entry. What does the match most directly indicate?
    • It identifies the person who uploaded the image
    • It proves every element of a criminal offense
    • The image is sufficiently similar to a known reference under the matching system
    • It establishes when the image was created
  2. How does known-content hash matching differ from a learned image classifier?
    • Hash matching compares fingerprints with reference material; a classifier estimates categories from learned patterns
    • Both methods require an exact byte-for-byte file copy
    • A classifier can only identify files already in a database
    • Hash matching determines who created an image
  3. A classifier flags an unfamiliar image with a high score. What should follow?
    • Automatic identification of the uploader
    • Public posting so others can vote
    • Treating the score as a court finding
    • A trained review under authorized procedures before consequential escalation
  4. Why may a perceptual-hash system still miss altered material?
    • The reference list contains only video
    • A substantial transformation may change the fingerprint beyond the system’s matching tolerance
    • Hash systems always require face recognition
    • Every image edit improves match certainty
  5. Why should a service distinguish a reference match from a classifier alert in its logs?
    • A single label makes performance easier to measure
    • The distinction changes the image’s file size
    • They have different evidentiary meaning and require different review paths
    • Both signals establish the same legal conclusion
  6. Which safeguard best protects sensitive material during review?
    • Allow every moderator to download flagged files
    • Store copies indefinitely for convenience
    • Share samples on public forums for verification
    • Restrict access to trained personnel, log handling, and avoid unnecessary copies
  7. What do NCMEC’s reference hashes help participating services do?
    • Recognize previously reported images or videos that match the reference set
    • Identify every new form of online abuse automatically
    • Determine whether an account holder is guilty
    • Replace required legal reporting processes
  8. Which performance report is most useful for responsible operation?
    • Publish only a single overall score
    • Separate false positives, missed known matches, review delays, and reporting accuracy
    • Count only how many images were scanned
    • Report the model version without any outcomes