The AI Learning Hub Journal
◆ References

Checking This Course Against the Sources

Weighing what you read about AI in auditthe higher the rung, the more weight a claim carries on its own — weight increases upwardsRUNGWHAT IT IS, AND WHY IT SITS HEREEXAMPLES BEHIND THIS COURSEWEIGHTStandards and regulationbinding — everything below it is a readingThe instruments that bind the engagementwhether or not anyone summarises themaccurately. Readings drift; the text does not.The IAASB’s ISAs; PCAOB auditingstandards; the ISQM quality-managementstandards; the IESBA Code; national regimesRegulator and inspection guidancehow the standards meet real filesNot standards themselves, but the closestavailable statement of how they will beapplied to files like yours. Dates quickly.Inspection reports and findings;thematic reviews from national auditregulators on technology and AI in auditAcademic evidencedisinterested and method-transparent — and slowNo product to sell, and the method is onthe page. Describes populations, not yourengagement, and lags practice by years.The automation-bias literature;research on audit data analytics andtechnology adoption in audit practicePractitioner commentaryfast and concrete, but nobody checked itSays what deployment actually feels like,which no regulator writes down. Generalisesfrom one firm, and is reviewed by no one.Conference talks, firm newsletters,blog posts, and single-engagementexperience written up informallyVendor materialwritten by the party selling the answerThe only source for what a product actuallydoes, and the least disinterested thing youwill ever read about whether it does it well.Product pages, accuracy claims,demonstrations, and case studieswith no independent replicationClosest to you and not on this ladder: your firm’s methodology, its approved-tool list and its independence rules —the only sources that know your firm, your clients and your engagement, and the group that settles the most real questionsRead downwards to find what binds you; read upwards to check what you were toldThis course is not on the ladder — it is model-drafted orientation, edited but unreviewed, and it carries no authority at all
Every rung has a use — the ladder is about how much weight a claim carries on its own.

How This Course Was Made

This course was drafted by an AI model working to an editorial brief and then edited. It has not been reviewed by a practising auditor, an audit methodology owner or an inspector in any jurisdiction, and it claims no authority. Treat it as orientation: vocabulary, the shape of the arguments, and the questions worth putting to the people whose job it is to be right about them. It is not a basis for an audit procedure, a reliance decision, a documentation approach or a compliance position — each of those needs a source that carries responsibility for the answer, and a course page does not. That is why the lessons name standards bodies and instrument families but never quote a clause, and why they say where an answer is jurisdictional or firm-specific instead of supplying one. Where the course could not be sure of a claim, the claim is not here — the final slide says exactly which ones.

  • Model-drafted to an editorial brief and edited — not reviewed by a practising auditor or an inspector
  • Orientation only: vocabulary, the shape of the arguments, and the questions worth asking
  • Never a basis for an audit procedure, a reliance decision or a compliance position
  • Where an answer is jurisdictional or firm-specific, the course says so instead of guessing

The Source Ladder

When you read anything about AI in audit, including this course, the useful first question is which rung it came from. At the top sit the standards and the law: the IAASB's International Standards on Auditing, PCAOB auditing standards, the quality-management standards, the IESBA Code and national regimes — these bind engagements whether or not anyone has summarised them accurately. Below them sits regulator and inspection guidance: thematic reviews and inspection findings, not standards themselves, but the closest available statement of how the standards will be applied to real files. Below that, academic evidence — the automation-bias literature and the audit-technology research — disinterested and method-transparent, but general and slow. Then practitioner commentary, which is fast, concrete and reviewed by nobody. Vendor material sits last: the only source for what a product actually does, and the least disinterested thing you will ever read about whether it does it well.

  • Standards and law bind the engagement; everything beneath them is somebody's reading, and readings drift
  • Inspection guidance and thematic reviews are the nearest statement of how standards meet real files
  • Academic evidence is disinterested and slow; practitioner commentary is fast and reviewed by nobody
  • Vendor material is the only source for what a product does, and the least disinterested about how well

The Sources This Course Rests On

These are the anchors the lessons were written against, grouped by rung and named by issuing body and subject so you can read them rather than take this course's word for anything. They are deliberately unlinked, because links rot and the issuing body's own site is where the current text lives. They are also deliberately without clause numbers, because a model-drafted course quoting clause numbers would be exactly the fluent, source-shaped confidence this course has spent four modules warning about. Two caveats matter more than the list itself. Standards and regimes differ by jurisdiction, and nothing here says which instrument governs your engagement — that is a question for your firm's methodology, not for a course page. And the final group settles more real questions than all of the others together, because it is the only one that knows your firm, your clients and your engagement letters.

  • Standards: the IAASB's ISAs on evidence, estimates, risk identification and using the work of experts; ISQM 1 on firm-level quality management; the IESBA Code on confidentiality and independence; PCAOB auditing standards
  • Regulators and inspection: PCAOB inspection reports, and thematic work from the FRC and other national audit regulators on audit technology and the use of AI in audit
  • Evidence: the academic literature on automation bias, and the research literature on audit data analytics and on technology adoption in audit practice
  • Closest to you, and not on this page: your firm's methodology, its approved-tool list and its independence rules — the only sources that know your engagement

Rejected for Uncertainty, and Keeping Current

This is the course's most honest slide: the claims it declined to make because it could not be sure, listed specifically so a practitioner reviewer can treat them as a research agenda rather than a shrug. Each is a real question with a real answer somewhere — in a standard's current text, a regulator's current position or a firm's legal advice — that a model working from training data could not pin down responsibly. On keeping current: the standard-setters named in this module have active projects on technology and AI in audit, and inspection regimes are publishing their early findings now, so positions taken here can quietly date without anything being retracted. Check the issuing body's own text rather than any summary, this one included. And if you find an error here, report it — naming sources is an invitation to check them, not a claim to have got them right.

  • Declined: whether and when any auditing standard or regulator requires disclosure of AI use in the audit — to those charged with governance, in the auditor's report, or in the file; the current IAASB and PCAOB texts and your own regulator's position are where the answer lives
  • Declined: the precise documentation each framework requires when a tool performs the reading — inputs, versions, outputs, review steps; the answer differs between ISA and PCAOB territory and needs the current text of each, read for your framework
  • Declined: jurisdiction-specific rules on data residency and cross-border processing of audit evidence and client data; this needs local counsel and your firm's own data arrangements, and no general course can answer it
  • Declined: how inspection regimes currently treat AI reliance in live inspections, and where liability sits when a validated, firm-approved tool errs on an engagement — both are moving now, and only current inspection reports and legal advice can say

Prefer slides, quizzes, and saved progress? Read this lesson in the library — free, no sign-up.