CRISC domain - 20% of the exam

Information Technology and Security

Information Technology and Security is 20% of the CRISC exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

58
Practice questions
8
Objectives
12
Easy
46
Medium

Where people go wrong

  • Naming a security objective the scenario rules out

    The wrong answers pick an objective that the facts of the scenario exclude. Encryption makes confidentiality look relevant, yet an administrator who silently altered stored results with no record caused an integrity failure. A denial of service attack is built to exhaust a service, so it threatens availability rather than the secrecy or accuracy of the data behind it. A copied file of unchanged records that stayed available to staff is a confidentiality breach, because an unauthorised party read it. The objective that failed is the one the facts describe.

    See it in a sample question
  • Mixing up recovery point, recovery time and maximum tolerable downtime

    The wrong answers swap the recovery objectives. The recovery point objective sets the maximum period of data that may be lost, so backups are checked against it, while the recovery time objective and maximum tolerable downtime both measure outage duration. A nightly full backup cannot meet a fifteen-minute recovery point objective, and moving the tapes closer only shortens restoration. Replication that silently fails for days leaves the secondary site three days behind, which breaches a one-hour recovery point objective in practice rather than the recovery time objective.

    See it in a sample question
  • Expecting the course itself to change behaviour

    The wrong answers look to the course itself for behaviour change. Reading logs, quiz scores, and counts of sessions delivered and staff registered record consumption, recall or reach, while a falling click rate and a rising report rate in simulated phishing show what staff actually do. A longer session, a harder end test or an on-demand video still gives one exposure that decays within weeks, so short, regular touchpoints sustain the behaviour. Polished videos, course length and module counts matter less than senior leaders visibly following the same controls.

    See it in a sample question
  • Offering a security control for a privacy principle

    The wrong answers reach for a security control where the scenario describes a privacy principle. Encryption and tighter physical access protect tapes that are still kept indefinitely, so a retention schedule with secure disposal is what fixes the disposal stage. Masking leaves unneeded data collected, whereas data minimisation limits collection to the email address required. Encryption in transit does not make a cross-border transfer lawful; a transfer mechanism binding the receiver does. Reusing claims data for marketing breaches purpose limitation, not confidentiality.

    See it in a sample question
  • Responding after the decision instead of before it

    The wrong answers place the risk response after the decision it should shape. Recording a tool after deployment locks in the redundancy, so architecture review belongs before approval. A post-build penetration test cannot keep pace with daily releases, and production monitoring finds a weakness only once exploited, so automated checks sit in each pipeline stage. Reviewing a generative AI trial afterwards leaves exposure unmanaged, so the practitioner first identifies the use case, data and exposures. Briefing governance only after an incident denies decision makers the chance to act.

    See it in a sample question

Objectives in this domain

Sample questions from this domain

Free sampleInformation Technology and Securitymedium

A hospital encrypts its patient records so that only authorised clinicians can read them. During an incident review, a risk practitioner finds that an administrator silently altered several stored results and no record of the change exists. Which security objective has failed in this case?

  • AConfidentiality has failed, because the encryption protecting the patient records was clearly bypassed by the administrator during the change.
  • BAvailability has failed, because clinicians were temporarily unable to reach the patient records while the administrator was making the changes.
  • CNon-repudiation has failed, because the administrator could later deny having sent any message that instructed the records to be updated.
  • DIntegrity has failed, because stored data was changed without authorisation and no trace of the modification was retained for review. Correct
Integrity assures data remains accurate and that unauthorised changes are prevented or at least detectable through evidence such as logging. The confidentiality, integrity and availability triad separates secrecy, accuracy and access. Encryption preserved secrecy and the records stayed reachable, but a stored value was changed without authorisation and left no audit trail, so the accuracy assurance, integrity, is the objective that failed.

Why A is wrong: Confidentiality is tempting because encryption is mentioned, but the records stayed unread by outsiders, so secrecy held; the breakdown was the undetected alteration of data.

Why B is wrong: Availability sounds relevant, yet nothing in the scenario shows the records became unreachable; the problem was that their accuracy was changed without any trace.

Why C is wrong: Non-repudiation concerns proving who originated an action; the core failure here is that the data itself was altered and unverifiable, which is an integrity loss.

Why D is correct: Integrity assures that data stays accurate and that unauthorised changes are detectable; a silent, untracked alteration of stored results is a direct loss of integrity.

Free sampleInformation Technology and Securitymedium

A records platform has a documented recovery point objective of one hour, and the backup design specifies replication every hour to a secondary site. A disaster recovery test reveals that the replication job has been silently failing for several days, so the secondary site holds data that is three days old. A risk practitioner is judging what this test result reveals about the recovery point objective. Which conclusion is best supported?

  • AThe recovery point objective is being breached in practice, because the secondary site can only restore three-day-old data despite a one-hour tolerable-loss target Correct
  • BThe recovery point objective is too strict and should be relaxed to three days to reflect what the replication arrangement is actually able to deliver in practice
  • CThe objective remains satisfied because the hourly replication frequency was correctly designed, and an occasional job failure does not affect the stated objective
  • DThe test mainly demonstrates a recovery time objective problem, since the delay in noticing the failed job lengthens how long restoration of the platform will take
A recovery point objective is met only when backup or replication controls actually operate, not merely when they are correctly designed. The recovery point objective caps how much data may be lost, judged against the freshness of recoverable data; a silently failed replication job leaves the secondary site days behind, so even though the design was sound the objective is breached in practice and needs monitoring of actual replication status.

Why A is correct: The objective is met only if recoverable data is no more than an hour old; with replication failing silently and the secondary site three days behind, actual recoverable data far exceeds the tolerable loss, so the objective is breached in practice.

Why B is wrong: Relaxing the objective to fit a broken control inverts the logic; the objective expresses the tolerable data loss the business has accepted and is not changed by a failed replication job.

Why C is wrong: A sound design is necessary but not sufficient; the objective is met only when the control actually operates, and a multi-day replication failure means the objective is being breached regardless of the design.

Why D is wrong: This conflates the two objectives; the stale data is a data-loss issue affecting the recovery point objective, not the duration of restoration that the recovery time objective measures.

Free sampleInformation Technology and Securityeasy

An organisation delivers a single security awareness session once a year and finds that risk-relevant behaviour, such as reporting suspicious emails, improves briefly after the session and then drops back within a few weeks. The risk practitioner is asked what change to the programme would best sustain the improved behaviour. What should the practitioner recommend?

  • AExtend the single annual session to a full day so that staff cover more material in one sitting and the learning is therefore retained for the rest of the year
  • BReinforce the key messages through short, regular touchpoints spread across the year rather than relying on the single annual session to carry the behaviour Correct
  • CAdd a longer final examination to the annual session and require staff to score highly before their attendance is formally recorded as complete
  • DReplace the live session with a recorded video that staff can watch at any point during the year whenever it happens to suit their own workload
Understand that spaced reinforcement across the year sustains awareness-driven behaviour better than a single annual event, however long or rigorous. Awareness gained from a one-off session decays quickly because the message is not refreshed, so the durable fix is spaced reinforcement through frequent short touchpoints that revisit the behaviour before it fades, whereas lengthening, testing, or recording the single event does not address the underlying decay.

Why A is wrong: A longer single session looks like more learning, but cramming more content into one event worsens fatigue and does not stop the rapid decay that already follows the session.

Why B is correct: Spaced repetition keeps the message present in daily working life, so behaviour is refreshed before it decays rather than relying on one event that fades within weeks.

Why C is wrong: A harder end test feels rigorous and raises measured scores, yet it proves recall on the day and does nothing to keep behaviour alive in the weeks that follow.

Why D is wrong: On-demand video is convenient and flexible, but a one-off viewing at a self-chosen time still gives a single exposure, so the same decay in behaviour will occur.

Free sampleInformation Technology and Securitymedium

A retailer is designing a new loyalty application and the project team proposes collecting customers' dates of birth, home addresses and income brackets even though only an email address is needed to operate the scheme. A risk practitioner is reviewing the design against data privacy principles. What should the practitioner recommend regarding the data being collected?

  • ALimit collection to the email address required to run the scheme, applying data minimisation so only data necessary for the purpose is gathered. Correct
  • BAllow all of the proposed fields to be collected now because gathering extra attributes early may support future analytics the business has not yet defined.
  • CCollect every proposed field but mask the more sensitive ones in the user interface so that frontline staff cannot view them on screen.
  • DCollect the full set of fields but obtain a broad blanket consent at sign-up that covers any present or future processing the retailer may wish to perform.
Data minimisation requires collecting only the personal data necessary for the stated purpose, so unneeded fields should not be gathered. Data minimisation limits collection to what is adequate, relevant and necessary for the declared purpose. Because the loyalty scheme functions with only an email address, gathering dates of birth, addresses and income creates needless privacy exposure, so the design should be pared back to the necessary field.

Why A is correct: Data minimisation requires collecting only what is necessary for the stated purpose, so restricting the design to the email address the scheme actually needs is the correct recommendation.

Why B is wrong: Collecting data for undefined future use is exactly what data minimisation forbids, so this option restates the flawed proposal rather than correcting it.

Why C is wrong: Masking limits who sees the data and sounds protective, but the data is still gathered and stored unnecessarily, which leaves the minimisation breach in place.

Why D is wrong: A blanket consent feels like it grants permission, but consent does not justify gathering data beyond what the purpose needs, so minimisation is still breached.

Free sampleInformation Technology and Securitymedium

A business unit at a retail bank wants to procure a specialist analytics platform that duplicates capabilities already provided by an existing enterprise data warehouse. A risk practitioner is asked how the enterprise architecture function should be used to manage the IT risk this decision creates. What is the most appropriate role for enterprise architecture here?

  • AUse the architecture review to assess the proposed platform against the target-state architecture and surface the redundancy and integration risk before approval. Correct
  • BLet the business unit procure the platform first and ask the architecture function to document it in the repository once it has been deployed into production.
  • CDefer entirely to the business unit because tool selection is a local operational matter that sits outside the remit of the enterprise architecture function.
  • DEscalate the request straight to the audit committee so an independent review can decide whether the analytics platform should be purchased at all.
Enterprise architecture review evaluates proposals against the target-state architecture to surface duplication and integration risk before approval. Enterprise architecture manages technology risk by comparing each proposed change to the agreed target state. Reviewing the analytics platform against that target reveals redundancy with the warehouse and the integration burden, allowing the risk to be addressed before the purchase rather than after deployment.

Why A is correct: Architecture review checks a proposal against the agreed target state, which is exactly where duplication and integration risk are caught before money is committed and the estate fragments.

Why B is wrong: Recording a tool after deployment keeps the repository current, but it does nothing to manage the risk of the purchase itself and locks in the redundancy the practitioner was asked about.

Why C is wrong: Treating selection as purely local is tempting for autonomy, but it ignores that overlapping platforms raise enterprise integration and cost risk that architecture exists to govern.

Why D is wrong: Audit-committee escalation sounds rigorous, but it bypasses the architecture review that is the correct first-line mechanism and is disproportionate to a routine procurement decision.

Other domains in this exam

See also the CRISC cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by ISACA. Original, blueprint-aligned practice material only.