CISSP Asset Security Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Data masking

  Why it's wrong here

  Data masking involves replacing sensitive, real data with realistic but fictitious data, typically for non-production environments like testing, development, or training. While it preserves the data's format and referential integrity, allowing applications to function correctly, its primary purpose is to prevent the exposure of actual sensitive information rather than to enable privacy-preserving statistical analysis through noise injection. It focuses on substitution to create safe copies, not on mathematically guaranteed privacy for aggregate queries.

• ✗

  Tokenization

  Why it's wrong here

  Tokenization is the process of replacing sensitive data elements with non-sensitive substitutes, known as tokens, which are typically randomly generated values. These tokens bear no mathematical or algorithmic relationship to the original data, making it extremely difficult to reverse-engineer the original value from the token. This technique is primarily used to reduce the scope of compliance requirements, such as PCI DSS, by ensuring sensitive data is not stored or processed directly, but it does not involve adding noise for privacy-preserving analytics.

• ✓

  Differential privacy

  Why this is correct

  Differential privacy is a rigorous mathematical framework that quantifies and limits the privacy risk to individuals when their data is part of a dataset used for statistical queries. It achieves this by strategically injecting calibrated noise into query results or the data itself, ensuring that the presence or absence of any single individual's data in the dataset does not significantly alter the output of an analysis. This allows for aggregate insights while providing strong, provable guarantees against re-identification, even by an attacker with auxiliary information.

• ✗

  Anonymization

  Why it's wrong here

  Anonymization involves removing or obscuring direct and indirect identifiers from a dataset to prevent the re-identification of individuals. Techniques include generalization (e.g., replacing exact age with age range), suppression (removing specific records or fields), and pseudonymization (replacing direct identifiers with pseudonyms). While it reduces re-identification risk, traditional anonymization often sacrifices data utility and does not offer the strong, quantifiable privacy guarantees against sophisticated inference attacks that differential privacy provides through the addition of noise.

Go deeper

Related to this question