ISC · Security and Confidentiality and Privacy · 10 practice questions
CPA Privacy: Are Tokenized Data Still Personal Information?
Tokenized data remain personal information when the holder can reasonably link them to identifiable individuals. Below, compare the same dataset as one fact changes, then practice identifying what controls the privacy conclusion.
Try one first
Hint
Focus on what makes data 'personal' for privacy rules, can the dataset be effectively linked back to individuals despite removing names?
Answer B. Privacy obligations attach to personal information, data that identifies or is reasonably linkable to an individual. Although direct identifiers were removed, retaining a separate token key that allows relinking means the dataset remains reidentifiable; therefore privacy-related requirements for collection, use, retention, and disclosure apply in addition to confidentiality controls.
Why not A: This distractor tempts because protecting commercially sensitive information is important, but competitive or commercial harm is primarily a confidentiality/business-risk concern. Privacy duties are triggered by whether data is personal/identifiable, not by whether its disclosure would hurt the company's competitive position.
Why not C: This is tempting because access controls reduce risk, but strong confidentiality protections do not change whether data is personal. If the dataset can be linked back to individuals, privacy obligations apply regardless of whether access is limited.
Why not D: This distractor is plausible because intended use can affect certain compliance obligations, but internal use alone does not eliminate privacy responsibilities. The primary question is identifiability, if reidentification is possible, privacy rules govern collection, use, retention, and disclosure irrespective of whether the firm plans to sell the data.
Same scenario, one fact changes
Base case
North Peak reviews an encrypted purchase-history dataset with persistent customer tokens but no names or emails. North Peak retains a separate lookup, the only reasonable way to identify customers. Its privacy notice covers personal data use and retention. All nonpublic data is labeled confidential. Assume no sector-specific law changes the analysis.
Answer: Yes. The dataset is personal information for North Peak.
North Peak can use its lookup to identify customers. Removing direct identifiers does not eliminate that link. Privacy analysis applies in addition to confidentiality controls.
Before you open each one, predict the answer.
Change 1An independent third party exclusively holds the lookup; North Peak cannot access or obtain it.
Answer: No. The dataset is not personal information for North Peak under these facts.
The lookup was the only reasonable identification route. Without access to it, North Peak cannot reasonably identify customers.
Change 2The lookup has been permanently destroyed, with no recoverable copies.
Answer: No. The dataset is not personal information for North Peak under these facts.
Destroying the only reasonable identification route removes North Peak's ability to link tokens to identified customers. Persistent tokens alone do not establish identifiability.
Change 3The dataset is unencrypted rather than encrypted.
Answer: Yes. The dataset remains personal information for North Peak.
Encryption does not decide identifiability. North Peak still holds the lookup, so this tempting change does not alter the privacy classification.
Key points
- Remember: A token hides a name, but an accessible lookup can make the record personal information.
- Evaluate linkability for North Peak, not merely for someone who holds a different set of information.
- Confidentiality labels and encryption do not determine whether records identify people.
- Personal data covered by North Peak's notice require privacy analysis of use and retention, not just protection from outsiders.
How the exam traps you
- Treating removal of names and emails as proof that records are not personal. Check whether North Peak can reconnect tokens to customers.
- Treating any persistent token as proof that North Peak can identify a customer. Look for a reasonably available lookup or another identification route.
- Assuming encryption or a confidential label eliminates privacy obligations. Separate protection of the dataset from its linkability to individuals.
Now the same facts as questions
Each question changes one fact from the one before. Watch which change flips the answer.
Question 1
Answer A. Correct. The retained lookup links tokens to identifiable customers.
Why not B: Incorrect. North Peak holds the lookup and can identify customers.
Why not C: Incorrect. Removing direct identifiers does not eliminate the lookup's link.
Why not D: Incorrect. A confidentiality label does not establish that information identifies individuals.
Question 2
Answer B. Correct. The inaccessible lookup leaves North Peak without a reasonable way to identify customers.
Why not A: Incorrect. North Peak cannot access the only reasonable identification route.
Why not C: Incorrect reasoning. Missing direct identifiers alone does not establish that reidentification is impossible.
Why not D: Incorrect. Confidential information need not be personal information.
Question 3
Answer B. Correct. No lookup or other reasonable identification route remains.
Why not A: Incorrect. The only reasonable identification route no longer exists.
Why not C: Incorrect reasoning. Destruction of the only identification route, not merely removing names, changes the answer.
Why not D: Incorrect. Confidential classification does not make otherwise nonidentifiable records personal.
Question 4
Answer A. Correct. North Peak still holds the lookup; encryption does not control linkability.
Why not B: Incorrect. The lookup remains available to North Peak.
Why not C: Incorrect. Tokens remain linked to customers through the separate lookup.
Why not D: Incorrect reasoning. Linkability, not the confidentiality label, makes these records personal.
Question 2
Hint
Focus on what distinguishes privacy-sensitive information from other confidential business information.
Answer A. Privacy concerns center on information about identified or reasonably identifiable individuals; privacy obligations govern how personal data are collected, used, retained, and disclosed. Confidentiality is broader and focuses on preventing unauthorized access to nonpublic information, including commercial documents that are not personal. Here, the employee leave-request forms include personal health details tied to named employees, so they raise privacy issues beyond mere confidentiality.
Why not B: This is tempting because confidentiality often depends on the potential business harm from disclosure, but it addresses commercial sensitivity rather than whether the data are personal and thus subject to privacy considerations.
Why not C: Limited access is an important safeguard for both confidential and private information and may seem relevant, but it is a control rather than the defining factor that makes information a privacy concern.
Why not D: Storage method affects security and operational risk but does not determine whether information is personal. Privacy status depends on the nature of the information and whether it is linked to an individual.
Question 3
Hint
First classify each dataset: is it personal information about identifiable individuals (privacy) or nonpersonal business information designated confidential under contract (confidentiality)?
Answer C. Privacy protections apply to personal information about identifiable individuals (names, SSNs, bank account numbers) and therefore primarily govern the employee data. Confidentiality covers nonpersonal proprietary business information that contracts designate as confidential, so the pricing formulas and product plans primarily implicate contractual confidentiality obligations. Receiving personal data from a client does not eliminate the privacy dimension.
Why not A: Candidates who think privacy only applies to data collected directly from individuals may choose this. That is incorrect: privacy obligations apply to personal information regardless of whether it was provided by a client on behalf of the individuals.
Why not B: This is tempting because both kinds of information deserve protection and contracts may impose similar controls, but it is wrong because privacy law is tied to personal information; nonpersonal business data does not, by itself, trigger privacy obligations even if it is sensitive.
Why not D: This appeals to the idea that any sensitive nonpublic information is a privacy issue, but privacy specifically concerns personal data about identifiable individuals; sensitive nonpersonal business information is a confidentiality issue under contract.
Question 4
Hint
Focus on what makes information a privacy matter, not on where it is stored or how it is protected.
Answer B. Privacy concerns arise primarily when information relates to an identified or reasonably identifiable person and is subject to limits on collection, use, retention, or disclosure. Confidentiality focuses on preventing unauthorized access to sensitive information more broadly (including trade secrets). Thus customer records can raise both privacy and confidentiality issues, while a pricing algorithm generally raises confidentiality concerns without being personal data.
Why not A: This is tempting because harm from disclosure is a classic reason to classify information as confidential. However, competitive harm is not the governing factor for privacy analysis. Privacy obligations are driven by the presence of personal data tied to individuals, not by whether the company would lose a business advantage.
Why not C: This choice is tempting because encryption is an important security control for sensitive data. But encryption is a protective measure, not the determinant of whether privacy obligations apply. Personal data remains subject to privacy requirements even when it is well encrypted.
Why not D: This distractor appeals to the common idea that outsourcing changes risk and responsibility. While storage location can affect control design and vendor oversight, it does not determine whether information is privacy-related. The key issue is the nature of the data, specifically whether it identifies a person.
Question 5
Hint
Think about whether the deciding factor is the nature of the data itself (personal information) or the presence of protections/agreements about secrecy.
Answer D. Privacy issues turn on whether the data are personal information and whether the entity has made commitments about how that personal information is collected, used, retained, disclosed, and disposed. Since the question is about retention and disposal, the primary determinant is whether the dataset is personal data subject to those privacy commitments; protections or business-confidential concerns are separate confidentiality matters.
Why not A: Contractual or regulatory retention rules are a plausible reason to keep or delete data, but such rules can apply to either personal (privacy) or business (confidentiality) data; they do not by themselves determine that the issue is a privacy matter.
Why not B: This option tempts because sensitivity labels and technical protections relate to how data are secured, but those attributes address confidentiality and security rather than whether retention/disposal is governed by privacy obligations tied to personal information.
Why not C: Competitive harm or NDA concerns indicate business-confidentiality issues that affect retention, but they do not make the matter one of privacy unless the data are personal information covered by privacy commitments.
Question 6
Hint
Ask what makes privacy different from ordinary protection of nonpublic business information.
Answer D. Privacy is triggered by information about identifiable individuals, especially when it includes personal or sensitive details (names, dates of birth, health enrollment). Confidentiality labels and commercial harm relate to information security and business confidentiality, but the essential privacy concern is the content and identifiability of the data.
Why not A: Where data is stored affects risk and controls, making this a plausible distraction. However, storage location does not determine whether information is a privacy matter, the deciding factor is whether the contents identify individuals.
Why not B: This is tempting because internal labels affect handling requirements, but an internal confidentiality label does not by itself make information a privacy issue; privacy depends on whether the data is personal and identifies individuals.
Why not C: This distractor plays on confidentiality concerns about competitive or reputational harm. While such harm justifies confidentiality protections, privacy is concerned primarily with risks to individuals from exposure of their personal information.
Common questions
Are tokenized customer records still personal information?
Yes, when the holder can reasonably reidentify customers. North Peak's separate lookup makes the base dataset personal information even without names or emails.
Is tokenized data personal information if another company holds the key?
The answer depends on reasonable linkability for the holder. In the variation, North Peak cannot access the third party's lookup and has no other reasonable identification route, so the records are not personal information for North Peak.
Does encrypting personal information remove privacy requirements?
No. Encryption protects data but does not change North Peak's ability to identify customers, so its privacy commitments about use and retention still matter.
Watch it solved
A full CPA ISC task-based simulation on Confidentiality and privacy, worked step by step.
Data Classification & Privacy Compliance | CPA ISC Simulation (TBS) on YouTube
Practice ISC like the real exam
The free ChatCPA simulator: real exam layout, timed testlets, starting with a question on this topic. No account needed to start.
Open the free simulator →More on Confidentiality and privacy
Questions from the ChatCPA bank of 17,658 CPA exam questions, each with a written reason for every wrong answer. ChatCPA is built by Nicholas Miller, CPA (Oregon #14907). How these pages are made. Spot an error? Tell us.