CDPSE Certified Data Privacy Solutions Engineer Questions and Answers

The use of personal data in model training is the primary privacy risk with LLMs, since once trained, models may retain, reproduce, or infer personal data without proper controls. Hallucinations (B), expertise shortages (C), and interoperability issues (D) are operational or performance risks, but not privacy risks.

“Training models on personal data can result in unintended retention, exposure, or disclosure of sensitive information.”

Privacy by design requires proactive, default, and continuous integration of privacy controls across the entire data life cycle (collection through disposal). Limiting to breach response (B) or remediation (C) is reactive, and focusing only on final product development (D) misses earlier phases where most risk originates.

“Embed privacy from the outset and across the full life cycle of processing activities.”

Senior leadership must define the roles and responsibilities of the person with oversight, who is responsible for ensuring compliance with the data privacy policy and applicable laws and regulations. This person may also be known as the data protection officer, the privacy officer, or the chief privacy officer, depending on the organization and jurisdiction. The person with oversight should have the authority, resources, and independence to perform their duties effectively.

D. Requiring employees to review the organization’s privacy policy on an annual basis

Short Explanation: Requiring employees to review the organization’s privacy policy on an annual basis is the best activity to enable an organization to identify gaps in its privacy posture because it can help to ensure that the employees are aware of the current privacy requirements, expectations, and practices of the organization. It can also help to identify any discrepancies, inconsistencies, or conflicts between the policy and the actual implementation of privacy controls and processes. By reviewing the policy regularly, the organization can also update and improve it as needed to reflect any changes in the privacy landscape, such as new laws, regulations, standards, or threats.

A retention schedule is a document that specifies how long different types of records or data should be kept and when they should be deleted or disposed of, based on legal, regulatory, operational or historical requirements. A retention schedule is the best indication of an effective records management program for personal data, as it reflects the principles of data minimization and storage limitation, which require limiting the collection, storage and processing of personal data to what is necessary and relevant for the intended purposes, and deleting or disposing of personal data when it is no longer needed or justified. A retention schedule also helps to reduce the privacy risks and costs associated with data storage and retention, such as data breaches, unauthorized access, misuse or loss of data. The other options are not as indicative of an effective records management program for personal data as a retention schedule. Archived data is used for future analytics may indicate that the organization is leveraging its data assets for business intelligence or research purposes, but it may not comply with the principles of data minimization and storage limitation, or the privacy rights and preferences of the data subjects. The legal department has approved the retention policy may indicate that the organization has obtained legal advice or guidance on its records management program for personal data, but it may not reflect the actual implementation or execution of the retention policy. All sensitive data has been tagged may indicate that the organization has implemented a data classification scheme for its records or data, but it may not indicate how long the records or data should be kept or when they should be deleted or disposed of1, p. 99-100 References: 1: CDPSE Review Manual (Digital Version)

The best answer is D. Mutual certificate authentication.

A comprehensive explanation is:

Mutual certificate authentication is a method of mutual authentication that uses public key certificates to verify the identities of both parties in a two-way communication. A public key certificate is a digital document that contains information about the identity of the certificate holder, such as their name, organization, domain name, etc., as well as their public key, which is used for encryption and digital signature. A public key certificate is issued and signed by a trusted authority, called a certificate authority (CA), that vouches for the validity of the certificate.

Mutual certificate authentication works as follows:

Both parties have a public key certificate issued by a CA that they trust.

When they initiate a communication, they exchange their certificates with each other.

They verify the signatures on the certificates using the CA’s public key, which they already have or can obtain from a trusted source.

They check that the certificates are not expired, revoked, or tampered with.

They extract the public keys from the certificates and use them to encrypt and decrypt messages or to generate and verify digital signatures.

They confirm that the identities in the certificates match their expectations and intentions.

By using mutual certificate authentication, both parties can be confident that they are communicating with the intended and legitimate party, and that their communication is secure and confidential.

Mutual certificate authentication is often used in conjunction with Transport Layer Security (TLS), a protocol that provides encryption and authentication for network communications. TLS supports both one-way and two-way authentication. In one-way authentication, only the server presents a certificate to the client, and the client verifies it. In two-way authentication, also known as mutual TLS or mTLS, both the server and the client present certificates to each other, and they both verify them. Mutual TLS is commonly used for secure web services, such as APIs or webhooks, that require both parties to authenticate each other.

Virtual private network (VPN), Secure Shell (SSH), and Transport Layer Security (TLS) are all technologies that can help to ensure the identities of individuals in a two-way communication are verified, but they are not methods of mutual authentication by themselves. They can use mutual certificate authentication as one of their options, but they can also use other methods, such as username and password, pre-shared keys, or tokens. Therefore, they are not as specific or accurate as mutual certificate authentication.

A privacy impact assessment (PIA) is a systematic process of identifying and evaluating the potential privacy risks and impacts of a data processing activity or system. A PIA helps to ensure that privacy is considered and integrated into the design and development of data processing activities or systems, and that privacy risks are mitigated or eliminated. A PIA also helps to determine the appropriate retention periods for personal data based on the purpose and necessity of the data processing, as well as the legal and regulatory obligations that apply to the data. Therefore, a PIA helps to define data retention time in a stream-fed data lake that includes personal data. References: : CDPSE Review Manual (Digital Version), page 99

Anonymization is a technique that removes or modifies all identifiers in a data set to prevent or limit the identification of the data subjects. Anonymization is an IT privacy practitioner’s best recommendation to reduce privacy risk before an organization provides personal data to a third party, as it would protect the privacy of the data subjects by reducing the linkability of the data set with their original identity, and also comply with the data minimization principle that requires limiting the collection, storage and processing of personal data to what is necessary and relevant for the intended purposes. Anonymization would also preserve some characteristics or patterns of the original data that can be used for analysis or research purposes by the third party, without compromising the accuracy or quality of the results. The other options are not as effective as anonymization in reducing privacy risk before an organization provides personal data to a third party. Tokenization is a technique that replaces sensitive or confidential data with non-sensitive tokens or placeholders that do not reveal the original data, but it does not prevent or limit the identification of the data subjects, as tokens can be reversed or linked back to the original data using a tokenization system or key. Aggregation is a technique that combines individual data into groups or categories that do not reveal the identity of the data subjects, but it may not prevent or limit the identification of the data subjects, as aggregated data can be de-aggregated or re-identified using other sources of information or techniques. Encryption is a technique that transforms plain text data into cipher text using an algorithm and a key, making it unreadable by unauthorized parties, but it does not prevent or limit the identification of the data subjects, as encrypted data can be decrypted or linked back to the original data using an encryption system or key1, p. 74-75 References: 1: CDPSE Review Manual (Digital Version)

Ensuring appropriate data classification should be done next after a migration of personal data involving a data source with outdated documentation has been approved by senior management, as it helps to identify the types, locations, and owners of the data, and to apply the appropriate privacy controls and measures based on the data classification level. Data classification also facilitates the data discovery, data minimization, data retention, and data disposal processes15. References: 1 Domain 3, Task 2; 5 Page 9

The practice that best indicates an organization follows the data minimization principle is that data is regularly reviewed for its relevance. The data minimization principle is one of the core principles of data protection under various laws and regulations, such as the GDPR or the CCPA. It states that personal data should be adequate, relevant, and limited to what is necessary in relation to the purposes for which they are processed. By regularly reviewing the data they hold, organizations can ensure that they do not collect or retain excessive or unnecessary data that may pose privacy risks or violate data subject rights.

Data is pseudonymized when being backed up, data is encrypted before storage, or data is only accessible on a need-to-know basis are also good practices for data protection, but they do not directly indicate that the organization follows the data minimization principle. Pseudonymization is a process of replacing identifying information in data with artificial identifiers or pseudonyms. Pseudonymization can help enhance the privacy of data by reducing the linkability between data and data subjects, but it does not prevent re-identification or inference attacks. Encryption is a process of transforming data into an unreadable form using a secret key or algorithm. Encryption can help protect the confidentiality, integrity, and availability of data by preventing unauthorized access, disclosure, or modification. Access control is a process of restricting who can access, modify, or delete data based on their roles, permissions, or credentials. Access control can help prevent unauthorized or inappropriate use of data by limiting the scope of access.

CDPSE/ISACA risk response taxonomy defines risk avoidance as deciding not to engage in the activity that gives rise to the risk. Reduction/mitigation (B/D) means proceed with controls; acceptance (C) means proceed without additional treatment. Not expanding is classic avoidance.

Key CDPSE-aligned phrasing (short extract): “Risk avoidance: Discontinue or do not initiate activities that create risk.”

Encryption of network traffic is the most important control to mitigate the risk of potential personal data compromise when employees access corporate IT assets remotely. Encryption is a process that transforms data into an unreadable form, making it difficult for unauthorized parties to intercept, modify, or steal it. Encryption of network traffic ensures that the data transmitted between the remote employees and the corporate network is protected from eavesdropping, tampering, or leakage.

Intrusion prevention system (IPS), firewall rules review, and intrusion detection system (IDS) are also useful controls for network security, but they are not as effective as encryption for protecting personal data in transit. IPS and IDS can monitor and block malicious or suspicious network traffic, but they cannot prevent data exposure if the traffic is intercepted by a third party. Firewall rules review can help optimize and secure the firewall configuration, but it cannot guarantee that the firewall will not be bypassed or compromised by an attacker. Therefore, encryption of network traffic is the best option among the choices given.

Automated data discovery and classification tools provide continuous and systematic identification of data assets across the enterprise, ensuring accuracy and reducing manual errors. Internal updates (B) and departmental reporting (D) depend on human diligence; periodic manual audits (C) are time-bound and may miss changes between reviews.

“Automated discovery ensures a current and reliable data inventory, critical for privacy compliance.”

The best way to address the concern that data collected by a third-party vendor and provided back to the organization may not be protected according to the organization’s privacy notice is to validate contract compliance. This means that the organization should verify that the third-party vendor is adhering to the terms and conditions of the contract, which should include clauses on data protection, privacy, and security. The contract should also specify the obligations and responsibilities of both parties regarding data collection, processing, storage, transfer, retention, and disposal. By validating contract compliance, the organization can ensure that the third-party vendor is following the same privacy standards and practices as the organization.

Asymmetric encryption is a method of encrypting and decrypting data using two different keys: a public key and a private key. The public key can be shared with anyone, while the private key is kept secret by the owner. Data encrypted with the public key can only be decrypted with the private key, and vice versa. Asymmetric encryption ensures the security of encryption keys when transferring data containing personal information between cloud applications, by providing the following benefits:

It can prevent unauthorized access or use of the data, as only the intended recipient who has the matching private key can decrypt the data sent by the sender who has the public key.

It can prevent man-in-the-middle attacks, where an attacker intercepts and modifies the data or keys in transit, as any tampering with the data or keys will result in decryption failure or error.

It can enable digital signatures, where the sender encrypts a message digest of the data with their private key, and the recipient verifies it with the sender’s public key. Digital signatures can ensure the authenticity and integrity of the data and the sender.

The other options are less effective or irrelevant for ensuring the security of encryption keys when transferring data containing personal information between cloud applications. Whole disk encryption is a method of encrypting all the data on a disk or device, such as a laptop or a smartphone. It does not protect the data when they are transferred over a network or stored on a cloud server. Symmetric encryption is a method of encrypting and decrypting data using the same key. It requires both parties to securely exchange and store the key, which may be difficult or risky in a cloud environment. Digital signature is not a method of encryption, but an application of asymmetric encryption that can provide additional security features for data transmission.

After a privacy risk has been accepted, the next step is to monitor the risk landscape for material changes. This means that the organization should keep track of any internal or external factors that may affect the likelihood or impact of the risk, such as new threats, vulnerabilities, regulations, technologies, or business processes. Monitoring the risk landscape can help the organization identify if the risk acceptance decision is still valid, or if it needs to be revisited or revised. Monitoring can also help the organization prepare for potential incidents or consequences that may arise from the accepted risk.

Reviewing proposed privacy rules that govern the processing of personal data is the most useful action to help define the scope of the project because it helps identify the legal and regulatory requirements, the data protection principles and the privacy objectives that the information security controls need to support. Reviewing recent audit reports, identifying databases that contain personal data or do not have encryption in place are helpful actions to assess the current state of privacy and security, but they do not provide a clear direction for the project scope.

Conducting a privacy impact assessment (PIA) should be done first to establish privacy by design when developing a contact-tracing application. A PIA is a systematic process that identifies and evaluates the potential effects of personal data processing operations on the privacy of individuals and the organization. A PIA helps to identify privacy risks and mitigation strategies at an early stage of development and ensures compliance with legal and regulatory requirements. Conducting a development environment review, identifying privacy controls, or identifying differential privacy techniques are important steps in privacy by design, but they should be done after conducting a PIA. References: CDPSE Exam Content Outline, Domain 2, Task 2.1

Prioritizing privacy-related risk scenarios as part of ERM processes is the best way to ensure that the risk responses meet the organizational objectives, because it helps to align the privacy risk management with the overall strategic goals, values, and culture of the organization. ERM is a holistic approach to identify, assess, and manage risks across the organization, taking into account the interdependencies and trade-offs among different types of risks. By integrating privacy-related risk scenarios into the ERM processes, the organization can evaluate the potential impact and likelihood of privacy risks on its mission, vision, and performance, and prioritize the most significant ones for mitigation or acceptance. This can also help to allocate appropriate resources, assign clear roles and responsibilities, and monitor and report on the effectiveness of the risk responses.

The strategic goals of the organization should be established first before a privacy office starts to develop a data protection and privacy awareness campaign, because they provide the direction, purpose, and scope of the campaign. The strategic goals of the organization reflect its vision, mission, values, and objectives, as well as its alignment with the relevant privacy laws and regulations, stakeholder expectations, and industry best practices. The privacy office should design and implement the awareness campaign in a way that supports and promotes the strategic goals of the organization, as well as measures and evaluates its effectiveness and impact.

To ensure the protection of personal data, privacy policies should mandate that access to information system applications be authorized by the business application owner, because they are the ones who are responsible for defining the business requirements, functions, and objectives of the applications. The business application owner can also determine the appropriate level of access for different users or groups based on their roles, responsibilities, and needs. The business application owner can also monitor and review the access control policies and procedures to ensure that they are effective and compliant with the privacy regulations and standards.

A thin client remote desktop protocol (RDP) is the most effective remote access model for reducing the likelihood of attacks originating from connecting devices, because it minimizes the amount of data and processing that occurs on the remote device. A thin client RDP only sends keyboard, mouse and display information between the remote device and the server, while the actual processing and storage of data happens on the server. This reduces the exposure of sensitive data and applications to potential attackers who may compromise the remote device.

Obtaining user consent is the best way to mitigate the privacy risk associated with setting cookies on a website. This means that the website should inform the users about the purpose, type, and duration of the cookies, and ask for their permission before storing or accessing any cookies on their browsers. This way, the users can exercise their right to control their personal data and opt-in or opt-out of cookies as they wish.

According to the General Data Protection Regulation (GDPR), consent must be freely given, specific, informed, and unambiguous. The website should provide clear and easy-to-understand information about the cookies and their implications for the users’ privacy, and offer a simple and effective way for the users to indicate their consent or refusal. The website should also respect the users’ choice and allow them to withdraw their consent at any time.

Implementing impersonation, ensuring nonrepudiation, and applying data masking are not relevant or effective methods to mitigate the privacy risk associated with setting cookies on a website. Impersonation means accessing or using data on behalf of another user, which could violate their privacy and security. Nonrepudiation means providing proof of the origin, authenticity, and integrity of data, which does not address the issue of user consent or preference. Data masking means hiding or replacing sensitive data with fake or modified data, which does not prevent the storage or access of cookies on the user’s browser.

Data mapping is the process of defining how data elements from different sources are related, transformed, and transferred to a common destination. Data mapping is the first step when developing an application link because it helps to ensure that the data exchanged between the API and the third-party application is consistent, accurate, and compatible. Data mapping also helps to identify any gaps, errors, or conflicts in the data and resolve them before the data transfer occurs.

Synthetic data generation is a recognized privacy-enhancing technology (PET) because it allows realistic model training and analysis without exposing actual personal data. PKI (A) provides authentication, not privacy preservation; blockchain (B) increases transparency but may conflict with privacy; identity management (D) supports security but is not a PET by itself.

“Synthetic data preserves patterns while removing identifiable personal information, enabling safe processing.”

Domain name system (DNS) sinkhole is a technology that redirects malicious or unwanted domain names to alternative destinations, such as a fake or harmless website, a warning page, or a null address. DNS sinkhole is the most effective technology deployed at an enterprise level to block malicious tracking of user internet browsing, as it would prevent users from accessing websites that use tracking technologies, such as cookies, web beacons, or fingerprinting, to collect and analyze user behavior or preferences. DNS sinkhole would also protect users from other malicious activities, such as malware distribution, phishing attempts, or botnet command and control. The other options are not as effective as DNS sinkhole in blocking malicious tracking of user internet browsing at an enterprise level. Web application firewall (WAF) is a technology that monitors and filters incoming and outgoing web traffic to protect web applications from attacks, such as cross-site scripting (XSS), SQL injection, or denial-of-service (DoS), but it does not block malicious tracking of user internet browsing. Website URL blacklisting is a method of blocking access to websites that are known or suspected to be malicious or harmful, but it does not block malicious tracking of user internet browsing from unknown or legitimate websites that use tracking technologies. Desktop antivirus software is a technology that scans and removes viruses, malware, spyware, or other threats from desktop computers or devices, but it does not block malicious tracking of user internet browsing from websites that use tracking technologies1, p. 92 References: 1: CDPSE Review Manual (Digital Version)

Differential privacy provides mathematically provable protection against re-identification and linkage, adding calibrated noise to outputs so individuals cannot be singled out or linked. Pseudonymization (C) and masking (D) reduce direct identifiers but remain vulnerable to linkage attacks. Secure multiparty computation (B) protects computation among parties, not release-time linkability.

“Differential privacy limits what can be learned about any individual from query results, resisting re-identification/linkage.”

A privacy impact assessment (PIA) is a systematic process to identify and evaluate the potential privacy impacts of a system, project, program or initiative that involves the collection, use, disclosure or retention of personal data. A PIA should be done first when planning a new implementation for tracking consumer web browser activity, as it would help to ensure that privacy risks are identified and mitigated before the implementation is executed. A PIA would also help to ensure compliance with privacy principles, laws and regulations, and alignment with consumer expectations and preferences. The other options are not as important as conducting a PIA when planning a new implementation for tracking consumer web browser activity. Seeking approval from regulatory authorities may be required or advisable for some types of personal data or data processing activities, but it may not be necessary or sufficient for tracking consumer web browser activity, depending on the context and jurisdiction. Obtaining consent from the organization’s clients may be required or advisable for some types of personal data or data processing activities, but it may not be necessary or sufficient for tracking consumer web browser activity, depending on the context and jurisdiction. Reviewing and updating the cookie policy may be required or advisable for some types of personal data or data processing activities, but it may not be necessary or sufficient for tracking consumer web browser activity, depending on the context and jurisdiction1, p. 67 References: 1: CDPSE Review Manual (Digital Version)

The right not to be profiled is the right of data subjects to not be subject to a decision based solely on automated processing, including profiling, which produces legal or similarly significant effects on them. The online application that delivers a mortgage decision automatically based on criteria set by the lender is an example of such a decision, as it affects the data subject’s ability to obtain a loan.

A PIA is a systematic process to identify and evaluate the potential privacy impacts of a system, project, program or initiative that involves the collection, use, disclosure or retention of personal data. A PIA should be conducted as early as possible in the system lifecycle, preferably during the development stage, to ensure that privacy risks are identified and mitigated before the system is deployed. Conducting a PIA during functional testing, UAT or production stages may be too late to address privacy issues effectively and may result in costly rework or delays1, p. 67 References: 1: CDPSE Review Manual (Digital Version)

The data controller determines the purposes and means of processing and therefore retains primary accountability for risk—even if a processor or custodian executes processing. Data processors (C) act under instructions; custodians (A) safeguard data but do not define processing; data scientists (B) are operational users, not accountable owners.

“The data controller remains accountable for compliance and risk regardless of outsourcing processing activities.”

Before an organization can respond to data subject access requests (DSARs), it needs to have a clear understanding of the data in its possession, such as what types of personal data are collected, where they are stored, how they are processed, who has access to them, and how long they are retained. This will help the organization to locate and retrieve the relevant data for each DSAR, and to ensure that the data are accurate, complete and up to date. Understanding the data in its possession will also help the organization to comply with other data protection principles and obligations, such as data minimization, purpose limitation, security and accountability.

The other options are less important or irrelevant to do first. Investing in a platform to automate data review may help to speed up the response process, but it does not guarantee that the organization has identified all the data sources and categories that are subject to DSARs. Confirming what is required for disclosure is also important, but it depends on the specific request and the applicable law or regulation. Creating a policy for handling access requests is a good practice, but it should be based on a thorough understanding of the data in its possession.

Anonymization is a technique that removes or modifies all identifiers in a data set to prevent or limit the identification of the data subjects. Anonymization is the IT privacy practitioner’s best recommendation for an organization that uses analytics derived from archived transaction data to create individual customer profiles for customizing product and service offerings, as it would protect the privacy of the customers by reducing the linkability of the data set with their original identity, and also comply with the data minimization principle that requires limiting the collection, storage and processing of personal data to what is necessary and relevant for the intended purposes. Anonymization would also preserve some characteristics or patterns of the original data that can be used for analysis or customization purposes, without compromising the accuracy or quality of the results. The other options are not as effective as anonymization in this situation. Discontinuing the creation of profiles is not a feasible or desirable option, as it would prevent the organization from achieving its business objectives and providing value to its customers. Implementing strong access controls is a security measure that restricts who can access, view or modify the data, but it does not address the issue of collecting or retaining more personal data than necessary or relevant. Encrypting data at rest is a security measure that transforms plain text data into cipher text using an algorithm and a key, making it unreadable by unauthorized parties, but it does not address the issue of collecting or retaining more personal data than necessary or relevant, and may require additional security measures to protect the encryption keys or certificates1, p. 75-76 References: 1: CDPSE Review Manual (Digital Version)

User consent to share personal data is the most important factor when designing APIs that enable mobile device applications to access personal data, as it ensures that the user is informed and agrees to the purpose, scope, and duration of the data sharing. User consent also helps to comply with the data protection principles and regulations, such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA), that require user consent for certain types of data processing and sharing134. References: 1 Domain 2, Task 7

The first consideration when selecting a data sanitization method is the type of storage device that holds the data to be sanitized. Different types of storage devices have different characteristics and limitations that affect the effectiveness and feasibility of data sanitization methods. For example, magnetic media, such as hard disk drives (HDDs), can be sanitized by data degaussing, which is wiping data permanently by weakening the magnetic field1. However, data degaussing is not applicable to devices that use solid state drive (SSD) technology, since SSDs do not store data magnetically2. Therefore, the storage type determines which data sanitization methods are suitable and available for the data disposal process.

The organization’s potential legal liabilities related to the data should be of greatest concern when an organization wants to store personal data in the cloud, as it may expose the organization to various compliance risks, such as data breach notification laws, data protection regulations, data sovereignty laws, and contractual obligations. The organization should ensure that the cloud storage provider complies with the applicable legal and regulatory requirements, and that the organization retains control and ownership of the data. The organization should also conduct due diligence and risk assessment of the cloud storage provider before entering into a contract. References: 2 Domain 2, Task 9; 4 

Privacy threat modeling is a methodology for identifying and mitigating privacy threats in a software architecture. It helps to ensure that privacy is considered in the design and development of software systems, and that privacy risks are minimized or eliminated. Privacy threat modeling typically involves the following steps: defining the scope and context of the system, identifying the data flows and data elements, identifying the privacy threats and their sources, assessing the impact and likelihood of the threats, and applying appropriate countermeasures to mitigate the threats. References: : CDPSE Review Manual (Digital Version), page 97

The best practice to protect data privacy when disposing removable backup media is B. Data sanitization.

A comprehensive explanation is:

Data sanitization is the process of permanently and irreversibly erasing or destroying the data on a storage device or media, such as a hard drive, a USB drive, a CD/DVD, etc. Data sanitization ensures that the data cannot be recovered or reconstructed by any means, even by using specialized software or hardware tools. Data sanitization is also known as data wiping, data erasure, data destruction, or data disposal.

Data sanitization is the best practice to protect data privacy when disposing removable backup media because it prevents unauthorized access, disclosure, theft, or misuse of the sensitive or confidential data that may be stored on the media. Data sanitization also helps to comply with the legal and regulatory requirements and standards for data protection and privacy, such as the General Data Protection Regulation (GDPR), the Health Insurance Portability and Accountability Act (HIPAA), the Payment Card Industry Data Security Standard (PCI DSS), etc.

There are different methods and techniques for data sanitization, depending on the type and format of the storage device or media. Some of the common methods are:

Overwriting: Overwriting replaces the existing data on the device or media with random or meaningless data, such as zeros, ones, or patterns. Overwriting can be done multiple times to increase the level of security and assurance. Overwriting is suitable for magnetic media, such as hard disk drives (HDDs) or tapes.

Degaussing: Degaussing exposes the device or media to a strong magnetic field that disrupts and destroys the magnetic structure and alignment of the data. Degaussing renders the device or media unusable and unreadable. Degaussing is suitable for magnetic media, such as hard disk drives (HDDs) or tapes.

Physical Destruction: Physical destruction involves applying physical force or damage to the device or media that breaks it into small pieces or shreds it. Physical destruction can be done by using mechanical tools, such as shredders, crushers, drills, hammers, etc., or by using thermal methods, such as incineration, melting, etc. Physical destruction is suitable for any type of media, such as hard disk drives (HDDs), solid state drives (SSDs), USB drives, CDs/DVDs, etc.

Data encryption (A) is not a good practice to protect data privacy when disposing removable backup media because it does not erase or destroy the data on the media. Data encryption only transforms the data into an unreadable format that can only be accessed with a key or a password. However, if the key or password is lost, stolen, compromised, or guessed by an attacker, the data can still be decrypted and exposed. Data encryption is more suitable for protecting data in transit or at rest, but not for disposing data.

Data scrambling © is not a good practice to protect data privacy when disposing removable backup media because it does not erase or destroy the data on the media. Data scrambling only rearranges the order of the bits or bytes of the data to make it appear random or meaningless. However, if the algorithm or pattern of scrambling is known or discovered by an attacker, the data can still be unscrambled and restored. Data scrambling is more suitable for obfuscating data for testing or debugging purposes, but not for disposing data.

Data masking (D) is not a good practice to protect data privacy when disposing removable backup media because it does not erase or destroy the data on the media. Data masking only replaces some parts of the data with fictitious or anonymized values to hide its true identity or meaning. However, if the original data is still stored somewhere else or if the masking technique is weak or reversible by an attacker, the data can still be unmasked and revealed. Data masking is more suitable for protecting data in use or in analysis, but not for disposing data.

The first consideration when conducting a privacy impact assessment (PIA) is the applicable privacy legislation that governs the collection, processing, storage, transfer, and disposal of personal data within the scope of the assessment. The applicable privacy legislation may vary depending on the jurisdiction, sector, or purpose of the data processing activity. The PIA should identify and comply with the relevant legal requirements and obligations for data protection and privacy, such as obtaining consent, providing notice, ensuring data quality and security, respecting data subject rights, and reporting data breaches. The applicable privacy legislation also determines the criteria, methodology, and documentation for conducting the PIA.

A privacy audit is a systematic and independent examination of an organization’s privacy policies, procedures, practices, and controls to assess their compliance with applicable laws, regulations, standards, and best practices. A privacy audit may result in various outputs, such as findings, recommendations, observations, or opinions. Among the options given, the output that is most likely to trigger remedial action is the identification of deficiencies in how personal data is shared with third parties. This is because such deficiencies may pose significant risks to the privacy and security of the data subjects, as well as to the reputation and legal liability of the organization. Remedial action may include implementing contractual safeguards, technical measures, or organizational changes to ensure that third parties respect and protect the personal data they receive from the organization.

The best approach for handling personal data that has been restricted is to flag users’ email addresses to make sure they do not receive promotional information, because this will respect the users’ preferences and rights to opt out of marketing communications. This will also help the company comply with the data protection laws and regulations that require consent and transparency for sending marketing emails, such as the General Data Protection Regulation (GDPR) and the CAN-SPAM Act12. The other options are not appropriate or sufficient for handling restricted data, because they may violate the users’ rights, expectations, or agreements, or cause operational issues for the company.

The scenario that poses the greatest risk to an organization from a privacy perspective is that the organization lacks a hardware disposal policy. A hardware disposal policy is a policy that defines how the organization should dispose of or destroy hardware devices that contain or process personal data, such as laptops, servers, hard drives, USBs, etc. A hardware disposal policy should ensure that personal data is securely erased or overwritten before the hardware device is discarded, recycled, donated, or sold. A hardware disposal policy should also comply with the applicable privacy regulations and standards that govern data retention and destruction. By lacking a hardware disposal policy, the organization exposes personal data to potential threats, such as theft, loss, or unauthorized access, use, disclosure, or transfer. References: : CDPSE Review Manual (Digital Version), page 123

The best way to safeguard the encryption keys is to ensure that they are stored in a cryptographic vault. A cryptographic vault is a secure hardware or software module that provides cryptographic services and protects the keys from unauthorized access, modification, or disclosure. A cryptographic vault can also provide other functions, such as key generation, key backup, key rotation, key destruction, and key auditing. A cryptographic vault can enhance the security and privacy of the encrypted data by preventing key compromise, leakage, or misuse. A cryptographic vault can also comply with the security standards and best practices for key management, such as the ISO/IEC 27002, NIST SP 800-57, or PCI DSS. References:

[ISACA Glossary of Terms]

[ISACA CDPSE Review Manual, Chapter 3, Section 3.3.3]

[ISACA Journal, Volume 4, 2019, “Key Management in the Multi-Cloud Environment”]

[ISACA CDPSE Review Manual, Chapter 3, Section 3.3.4]

Collecting only the data necessary to meet objectives is the best approach to minimize privacy risk when collecting personal data. This is based on the principle of data minimization, which states that personal data should be adequate, relevant, and limited to what is necessary in relation to the purposes for which they are processed. Using a third party, collecting data through a secure web server, or aggregating data immediately may reduce some privacy risks, but they do not eliminate the possibility of collecting excessive or unnecessary data. References: CDPSE Exam Content Outline, Domain 3, Task 3.2

Zero-knowledge proofs (ZKPs) allow a party to prove possession of a secret or attribute without revealing the secret itself, enabling privacy-preserving verification. Digital signatures (A) and PGP (B) still involve exposure of certain data elements/keys for validation; passwordless methods (D) change authentication factors, not zero-knowledge verification.

“Zero-knowledge proofs enable verification without disclosure of the underlying data.”

The most effective way to support organizational privacy awareness objectives is D. Customizing awareness training by business unit function.

A comprehensive explanation is:

Organizational privacy awareness objectives are the goals and expectations that an organization sets for its employees and stakeholders regarding the protection and management of personal data. Privacy awareness objectives may vary depending on the nature, scope, and purpose of the organization’s data processing activities, as well as the legal, regulatory, contractual, and ethical obligations and implications that apply to them.

One of the best practices to support organizational privacy awareness objectives is to customize awareness training by business unit function. This means that the organization should design and deliver privacy awareness training programs that are tailored to the specific roles, responsibilities, and needs of each business unit or department within the organization. Customizing awareness training by business unit function can have several benefits, such as:

Enhancing the relevance and effectiveness of the training content and methods for each audience group, by addressing their specific privacy challenges, risks, and opportunities.

Increasing the engagement and motivation of the trainees, by showing them how privacy relates to their daily tasks, goals, and performance.

Improving the retention and application of the training knowledge and skills, by providing practical examples, scenarios, and exercises that reflect the real-world situations and problems that the trainees may encounter.

Fostering a culture of privacy across the organization, by creating a common language and understanding of privacy concepts, principles, and practices among different business units or departments.

Some examples of how to customize awareness training by business unit function are:

Providing different levels or modules of training based on the degree of access or exposure to personal data that each business unit or department has. For example, a basic level of training for all employees, an intermediate level of training for employees who handle personal data occasionally or incidentally, and an advanced level of training for employees who handle personal data regularly or extensively.

Providing different topics or themes of training based on the type or category of personal data that each business unit or department processes. For example, a general topic of training for employees who process non-sensitive or non-personal data, a specific topic of training for employees who process sensitive or special data categories (such as health, biometric, financial, or political data), and a specialized topic of training for employees who process high-risk or high-value data (such as intellectual property, trade secrets, or customer loyalty data).

Providing different formats or modes of training based on the preferences or constraints of each business unit or department. For example, a face-to-face format of training for employees who work in the same location or office, an online format of training for employees who work remotely or across different time zones, and a blended format of training for employees who work in a hybrid mode or have flexible schedules.

The other options are not as effective as option D.

Funding in-depth training and awareness education for data privacy staff (A) may improve the competence and confidence of the data privacy staff who are responsible for designing and implementing the privacy policies and practices of the organization, but it does not necessarily support the organizational privacy awareness objectives for the rest of the employees and stakeholders.

Implementing an annual training certification process (B) may ensure that the employees and stakeholders are updated and refreshed on the privacy policies and practices of the organization on a regular basis, but it does not necessarily address their specific privacy needs and challenges based on their business unit function.

Including mandatory awareness training as part of performance evaluations © may incentivize the employees and stakeholders to participate in and complete the privacy awareness training programs offered by the organization, but it does not necessarily enhance their understanding and application of privacy concepts and principles based on their business unit function.

CDPSE emphasizes outcomes of training—measurable role-based understanding and behavior—over mere completion or absence of incidents. HR mandates (B) and clean audits (D) show activity or point-in-time results, not sustained effectiveness. No incidents (A) is not a reliable indicator of program quality.

Key CDPSE-aligned phrasing (short extract): “Effective awareness is role-based and demonstrated in behavior.”

Requiring independent audits of the providers’ data privacy controls is the best way to ensure third-party providers that process an organization’s personal data are addressed as part of the data privacy strategy. Independent audits can verify that the providers are complying with the applicable data privacy laws and regulations, as well as the organization’s own policies and standards. Independent audits can also identify any gaps or weaknesses in the providers’ data privacy controls and recommend corrective actions or improvements.

Obtaining executive support is the first step in developing an organization-wide strategy to address data privacy risk, as it ensures that the privacy program has the necessary resources, authority, and alignment with the organization’s goals and objectives. Without executive support, the privacy program may face challenges in implementing and enforcing privacy policies, procedures, and controls across the organization. References: 2 Domain 1, Task 1

Any organization collecting information about EU residents is required to operate with transparency in collecting and using their personal information. Chapter III of the GDPR defines eight data subject rights that have become foundational for other privacy regulations around the world:

Right to access personal data. Data subjects can access the data collected on them.

One of the privacy requirements related to the rights of data subjects is the right to access, which means that individuals have the right to obtain a copy of their personal data, as well as information about how their data is processed, by whom, for what purposes, and for how long. To meet this requirement, an organization’s technology stack should incorporate features that allow individuals to have direct access to their data, such as self-service portals, dashboards, or applications. This way, individuals can exercise their right to access without relying on intermediaries or manual processes, which can be inefficient, error-prone, or insecure. References: : CDPSE Review Manual (Digital Version), page 137

The primary means by which an organization communicates customer rights as it relates to the use of their personal information is publishing a privacy notice. A privacy notice is a document that informs the customers about how the organization collects, uses, shares, and protects their personal information, and what rights and choices they have regarding their data4. A privacy notice is a legal requirement under many data protection laws and regulations, such as the General Data Protection Regulation (GDPR), the California Consumer Privacy Act (CCPA), or the Personal Information Protection and Electronic Documents Act (PIPEDA)5 . A privacy notice is also a good practice to demonstrate the organization’s commitment to transparency, accountability, and customer trust. References:

ISACA Glossary of Terms

Article 13 and 14 of the GDPR

[Section 1798.100 of the CCPA]

[Schedule 1, Principle 4.8 of the PIPEDA]

[ISACA CDPSE Review Manual, Chapter 1, Section 1.3.2]

The first thing that an IT privacy practitioner should do before an organization migrates personal data from an on-premise solution to a cloud-hosted solution is to perform a privacy impact assessment (PIA). A PIA is a systematic process of identifying and evaluating the potential privacy risks and impacts of a data processing activity or system. A PIA helps to ensure that privacy is considered and integrated into the design and development of data processing activities or systems, and that privacy risks are mitigated or eliminated. A PIA also helps to determine the appropriate measures to protect personal data in a cloud-hosted solution, such as encryption, pseudonymization, anonymization, access control, audit trail, breach notification, etc. A PIA also helps to comply with the applicable privacy regulations and standards that govern data processing activities in a cloud-hosted solution. References: : CDPSE Review Manual (Digital Version), page 99

Application hardening measures are the most important action to protect a mobile banking app and its data against manipulation and disclosure because they prevent attackers from reverse engineering, tampering, or injecting malicious code into the app. Application hardening measures include techniques such as code obfuscation, encryption, integrity checks, anti-debugging, and anti-tampering mechanisms. These measures make the app more resilient and secure against various types of cyberattacks.

Validating the privacy framework is a responsibility of the audit function in helping an organization address privacy compliance requirements, as it would help to verify and validate the effectiveness and adequacy of the privacy framework implemented by the organization to comply with privacy principles, laws and regulations. Validating the privacy framework would also help to identify and report any gaps, weaknesses or issues in the privacy framework, and to provide recommendations for improvement or remediation. The other options are not responsibilities of the audit function in helping an organization address privacy compliance requirements. Approving privacy impact assessments (PIAs) is a responsibility of management or governance function in helping an organization address privacy compliance requirements, as they would have authority and accountability for approving PIAs conducted by project teams or business units before implementing any system, project, program or initiative that involves personal data processing activities. Managing privacy notices provided to customers is a responsibility of operational function in helping an organization address privacy compliance requirements, as they would have direct contact and interaction with customers and would be responsible for providing clear and accurate information about how their personal data is collected, used, disclosed and transferred by the organization. 

The best testing method to identify and review the application’s runtime modules is dynamic application security testing (DAST). DAST is a testing technique that analyzes the application’s behavior and functionality during its execution. DAST can detect security and privacy vulnerabilities that are not visible in the source code, such as injection attacks, cross-site scripting, broken authentication, sensitive data exposure, or improper error handling. DAST can also simulate real-world attacks and test the application’s response and resilience. DAST can provide a comprehensive and realistic assessment of the application’s security and privacy posture in the pre-production environment. References:

[ISACA Glossary of Terms]

[OWASP Top 10 Web Application Security Risks]

[ISACA CDPSE Review Manual, Chapter 2, Section 2.4.2]

[ISACA Journal, Volume 6, 2018, “Dynamic Application Security Testing”]

Authentication is a process of verifying the identity of a user or device that requests access to a system or resource. Authentication can be based on one or more factors, such as something the user knows (e.g., password), something the user has (e.g., token), something the user is (e.g., fingerprint) or something the user does (e.g., signature). When an organization requires a photo on a government-issued identification card to validate an in-person credit card purchase, it is using possession factor authentication, which relies on something the user has as proof of identity. The other options are not applicable in this scenario1, p. 81 References: 1: CDPSE Review Manual (Digital Version)

The primary reason to complete a privacy impact assessment (PIA) is to understand privacy risks associated with the collection, use, disclosure or retention of personal data. A PIA is a systematic process to identify and evaluate the potential privacy impacts of a system, project, program or initiative that involves personal data processing activities. A PIA helps to ensure that privacy risks are identified and mitigated before the implementation is executed. A PIA also helps to ensure compliance with privacy principles, laws and regulations, and alignment with customer expectations and preferences. The other options are not primary reasons to complete a PIA. To comply with consumer regulatory requirements may be a reason to complete a PIA, but it is not the primary reason, as consumer regulatory requirements may vary depending on the context and jurisdiction. To establish privacy breach response procedures may be an outcome of completing a PIA, but it is not the primary reason, as privacy breach response procedures are only one aspect of mitigating privacy risks. To classify personal data may be an activity that is part of completing a PIA, but it is not the primary reason, as personal data classification is only one aspect of understanding privacy risks1, p. 67 References: 1: CDPSE Review Manual (Digital Version)

Critical data elements are the data elements that are essential for the organization to achieve its business objectives, comply with legal and regulatory requirements, and protect the privacy and security of the data subjects. Critical data elements should be mapped to the data process flow, which is a graphical representation of how data is collected, processed, stored, shared, and disposed of within the organization. Mapping critical data elements to the data process flow helps to identify the sources, destinations, transformations, and dependencies of the data, as well as the potential risks and controls associated with each step of the data lifecycle.

Input validation controls are the best way to ensure consumer credit card numbers are accurately captured. Input validation controls are methods that check the format, type, range, and length of the input data before accepting, processing, or storing it. Input validation controls can help prevent errors, fraud, or data loss by rejecting invalid, incomplete, or malicious input. For example, input validation controls can verify that a credit card number follows the Luhn algorithm1, has the correct number of digits2, and matches the card issuer’s prefix3. Input validation controls can also prevent SQL injection attacks4 or cross-site scripting attacks5 that may compromise the security and privacy of the data.

Input reference controls, access controls, and reconciliation controls are also important for data quality and security, but they do not directly ensure the accuracy of consumer credit card numbers. Input reference controls are methods that compare the input data with a predefined list of values or a reference table to ensure consistency and validity. For example, input reference controls can check if a country name or a postal code is valid by looking up a database of valid values. Access controls are methods that restrict who can access, modify, or delete the data based on their roles, permissions, or credentials. For example, access controls can prevent unauthorized users from accessing or tampering with consumer credit card numbers. Reconciliation controls are methods that compare the data from different sources or systems to ensure completeness and accuracy. For example, reconciliation controls can check if the transactions recorded in the accounting system match the transactions processed by the payment gateway.

Data classification is the process of categorizing data according to its sensitivity, value, and criticality for the organization and the data subjects. Data classification has the greatest impact on the treatment of data within the scope of an organization’s privacy policy, as it determines the appropriate level of protection, access, retention, and disposal for each type of data. Data classification also helps to comply with the privacy principles and regulations, such as data minimization, purpose limitation, accuracy, security, and accountability.

The principle of data minimization states that personal data should be adequate, relevant and limited to what is necessary in relation to the purposes for which they are processed. By using only the data required by the application, the organization can reduce the amount of data that is collected, stored, processed and potentially exposed. This can also help the organization comply with privacy laws and regulations that require data minimization, such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA).

Sensitive personal data is a subset of personal data that reveals or relates to more intimate or confidential aspects of a person’s identity, such as their racial or ethnic origin, religious or philosophical beliefs, health status, sexual orientation, political opinions, trade union membership, biometric or genetic data, or criminal record. Sensitive personal data is subject to more stringent legal and regulatory protections and requires a higher level of consent from the data subject to be processed. Mailing address, bank account login ID, and contact phone number are examples of personal data, but not sensitive personal data, as they do not reveal or relate to such intimate or confidential aspects of a person’s identity.

A multinational organization that operates across different countries and regions should perform an annual review of changes to privacy regulations in all jurisdictions where its corporate data is processed. This is because different jurisdictions may have different privacy laws and requirements that apply to the collection, use, storage, transfer, and disposal of personal data. For example, the EU General Data Protection Regulation (GDPR) applies to any organization that processes personal data of individuals in the EU, regardless of where the organization is located or where the data is processed. Therefore, the organization should keep track of the changes to privacy regulations in all relevant jurisdictions and update its data privacy policy accordingly to ensure compliance and avoid penalties or lawsuits.

The primary means by which an organization communicates customer rights as it relates to the use of their personal information is publishing a privacy notice. A privacy notice is a document that informs the customers about how their personal information is collected, used, shared, stored, and protected by the organization, as well as what rights they have regarding their personal information, such as access, rectification, erasure, portability, objection, etc. A privacy notice should be clear, concise, transparent, and easily accessible to the customers, and should comply with the applicable privacy regulations and standards. A privacy notice helps to establish trust and transparency between the organization and the customers, and enables the customers to exercise their rights and choices over their personal information. References: : CDPSE Review Manual (Digital Version), page 39

The best course of action to prevent false positives from data loss prevention (DLP) tools is to re-establish baselines for configuration rules. False positives are events that are triggered by a DLP policy in error, meaning that the policy has mistakenly identified non-sensitive data as sensitive or blocked legitimate actions. False positives can reduce the effectiveness and efficiency of DLP tools by generating unnecessary alerts, wasting resources, disrupting workflows, and creating user frustration. To avoid false positives, DLP tools need to have accurate and updated configuration rules that define what constitutes sensitive data and what actions are allowed or prohibited. Configuration rules should be based on clear and consistent criteria, such as data classification levels, data sources, data destinations, data formats, data patterns, user roles, user behaviors, etc. Configuration rules should also be regularly reviewed and adjusted to reflect changes in business needs, regulatory requirements, or threat landscape.

Conducting additional discovery scans, suppressing the alerts generating the false positives, or evaluating new DLP tools are not the best ways to prevent false positives from DLP tools. Conducting additional discovery scans may help identify more sensitive data in the network, but it does not address the root cause of false positives, which is the misconfiguration of DLP policies. Suppressing the alerts generating the false positives may reduce the noise and annoyance caused by false positives, but it does not solve the problem of inaccurate or outdated DLP policies. Evaluating new DLP tools may offer some advantages in terms of features or performance, but it does not guarantee that false positives will be eliminated or reduced without proper configuration and tuning of DLP policies.