Which of the following methods does NOT contribute to keeping the data confidential?
Answer(s): D
Why option D is the correct choiceReferential integrity is a database-level constraint that guarantees that every foreign-key value points to an existing primary-key value. It ensures data consistency and accuracy, but it does nothing to hide, obfuscate, or protect the actual content of the data. Consequently, it does not contribute to the confidentiality of the information.Why the other options do enhance confidentialityDifferential privacy injects mathematically calibrated noise into query results, limiting the ability of an attacker to infer whether any single record is present in the dataset. This directly protects the privacy of individual subjects.Homomorphic encryption allows computations to be performed on ciphertexts, so data can be processed without ever being decrypted. The raw data remains encrypted end-to-end, preserving confidentiality while still enabling analytical operations. k-anonymity generalizes quasi-identifiers until each record is indistinguishable from at least k-1 others, preventing re-identification and thus shielding the underlying personal details from disclosure.In summary, options A, B, and C are privacy-preserving techniques that actively limit exposure of the underlying data, whereas referential integrity merely enforces structural correctness and offers no confidentiality benefit.
Differential Privacy – “The Differential Privacy Toolkit” (Google AI) – https://github.com/google/differential-privacy Homomorphic Encryption – “Introduction to Homomorphic Encryption” (Microsoft Research) – https://www.microsoft.com/en-us/research/project/homomorphic-encryption/ k-Anonymity – “k-Anonymity and its Extensions” (IEEE Security & Privacy) – https://ieeexplore.ieee.org/document/4534655 Referential Integrity – “Database Referential Integrity Constraints” (NIST SP 800-53 Rev. 5) – https://csrc.nist.gov/publications/detail/sp/800-53/rev-5 (Note: this source discusses integrity constraints, not confidentiality techniques.)
What element is most conducive to fostering a sound privacy by design culture in an organization?
Why option D is the best answerSenior-management advocacy creates the cultural foundation – When leaders actively champion privacy, they allocate budget, embed privacy criteria into project gates, and model the desired behaviours. This top-down commitment demonstrated in internal policy documents and performance metrics is the single driver that turns privacy principles into everyday practice, the hallmark of a “privacy-by-design” culture. It enables the other initiatives to succeed – Allocation of resources, authority to enforce privacy requirements, and the ability to influence cross-functional teams stem from leadership buy-in. Without executive backing, even comprehensive training or periodic policy reviews lack enforcement power and may be deprioritised.Why the other options are less suitableOption A – Policy acknowledgment – Appreciation of a privacy policy is a compliance checkbox; it does not guarantee that employees understand how to apply privacy concepts or that the organization embeds privacy into design decisions. Option B – Frequent awareness training – Training raises awareness but, without leadership sponsorship, its impact is limited to knowledge transfer rather than structural change. Training alone cannot compel teams to integrate privacy into architecture, processes, or product development. Option C – Monthly principle reviews – Periodic reviews can surface gaps, yet they are primarily retrospective and reactive. They lack the proactive, strategic influence needed to shape design choices and organisational norms unless backed by senior advocacy.Thus, gaining advocacy from senior management (Option D) most directly establishes the environment needed for a sound privacy-by-design culture.
International Association of Privacy Professionals (IAPP) – Privacy by Design: A Practical Guide: https://iapp.org/resources/privacy-by-design-a-practical-guide/ European Commission – Privacy by Design (official guidance): https://ec.europa.eu/info/law/law-topic/data-protection_en#privacy_by_design
After stringent testing an organization has launched a new web-facing ordering system for its consumer medical products. As the medical products could provide indicators of health conditions, the organization could further strengthen its privacy controls by deploying?
JustificationThe ordering system processes health-related data that can be used to infer sensitive medical conditions; therefore, the organization must limit the identifiability of individuals in any output or analytics generated from this data. Differential identifiability provides a mathematically proven guarantee that the release of analytical results (e.g., aggregated sales, trend reports) cannot be reverse-engineered to re-identify patients, even when combined with external datasets. This aligns with the privacy principle of “data minimisation” and “purpose limitation” emphasized in the CIPP/E and CIPP/US frameworks. Implementing differential privacy mechanisms also supports compliance with regulations that govern health information (e.g., HIPAA, GDPR health-data provisions) by reducing the risk of re-identification while still permitting useful insights.Why the other options are less appropriateA – Run-time behavior monitoring focuses on detecting anomalies or attacks at the application level; it does not address the core privacy risk of exposing health-condition indicators through processed outputs. B – Content delivery network (CDN) improves performance and availability of static assets; it adds no inherent privacy controls and could even increase exposure if not paired with access-control measures. C – Context-aware computing adapts functionality based on user context but does not inherently prevent the derivation of identifiable health patterns from aggregated data; it may even amplify privacy concerns if context includes sensitive attributes.Therefore, deploying differential identifiability mechanisms best strengthens privacy for this health-focused ordering platform.ReferencesIAPP CIPP/E Body of Knowledge – Differential Privacy Overview: https://iapp.org/resources/cipppe-body-of-knowledge/ NIST Special Publication 800-188 – Differential Privacy for the Public Release of Statistics: https://csrc.nist.gov/publications/detail/sp/800-188/finalAny other questions?
Machine-learning based solutions present a privacy risk because?
Answer(s): A
Technical justificationOption A – Training data compromisedThe core privacy risk in machine-learning pipelines is the exposure or contamination of training data, which can leak personal or sensitive information used to build the model. If the training set contains protected data, that data can be inferred or extracted through model inversion or membership attacks, directly violating data-subject confidentiality. This aligns with the CIPP/E principle that “personal data must be processed lawfully, fairly and in a way that ensures appropriate security,” and it is a concrete technical threat that can be mitigated by data-governance and de-identification practices.Option B – Inherent bias from developersBias is a fairness and ethical concern; it may lead to discriminatory outcomes, but it does not create a direct privacy breach. It impacts equity rather than confidentiality or data protection.Option C – Decision-making process not documentedLack of documentation affects model transparency and auditability, which is important for accountability, yet it does not by itself compromise personal data. The privacy risk here is about openness, not about unauthorized data exposure.Option D – More vulnerabilities than other softwareWhile ML models can introduce new attack surfaces, the statement is overly broad and not specifically tied to privacy. The primary privacy-related vulnerability remains the handling of training data, not a generic increase in overall vulnerabilities.Conclusion The most direct privacy risk unique to machine-learning solutions is the potential compromise of the training dataset, making Option A the correct answer.
IAPP CIPT Body of Knowledge – Training Data Privacy: https://iapp.org/resources/cipptraining/ NIST Special Publication 800-53 Rev. 5 – Privacy Controls for Machine Learning Systems: https://csrc.nist.gov/publications/detail/sp/800-53/rev-5/final
An organization is launching a new smart speaker to the market. The device will have the capability to play music and provide news and weather updates. Which of the following would be a concern from a privacy perspective?
Why D – “Context-aware computing” is the correct privacy concernA smart speaker that plays music, news, and weather continuously monitors audio input, usage patterns, preferred content, and ambient conditions (e.g., temperature, occupancy). This context-aware capability enables collection of highly detailed behavioral data without explicit user consent, making it a prime source of privacy risk. Privacy-by-design frameworks (e.g., NISTIR 7628, ISO/IEC 27701) specifically flag context-aware systems as needing safeguards such as purpose limitation, data minimisation, and transparent consent because they can infer sensitive attributes from seemingly benign data streams. The other options refer to unrelated concepts:Appropriation – concerns the unauthorized use of another’s property or assets, not applicable to a speaker’s functionality. Browser Fingerprinting – a tracking technique tied to web browsers, not to a dedicated voice-assistant device. Context of authority – relates to governance or jurisdictional authority, which does not directly describe the privacy implications of a smart speaker’s operation.Thus, the privacy issue stems from the device’s capacity to sense and act upon contextual cues, making Context-aware computing the most relevant concern.
International Association of Privacy Professionals (IAPP), Privacy by Design guide: https://iapp.org/resources/account/general/privacy-by-design/ NIST, Guide to Experimental and Testing Methods for Evaluating Privacy and Security in AI Systems (NISTIR 7628): https://www.nist.gov/itl/applied-cybersecurity/NISTIR7628
Post your Comments and Discuss IAPP CIPT exam prep with other Community members:
Our website is free, but we have to fight against AI bots and content theft. We're sorry for the inconvenience caused by these security measures. You can access the rest of the CIPT content, but please register or login to continue.