Quiz dotyczący oceny ryzyka audytu AI i planowania

Odpowiedzi i wyjaśnienia pojawiają się w miarę rozwiązywania quizu.

1. A journal entry model gives an entry its highest risk score. What does that score mean?
   • The entry is a confirmed misstatement
   • The entry must be reversed by the client
   • The entry is a reason to investigate, not a finding of misstatement
   • The entry is excluded from sampling
2. An account scores low on every analytic but is a significant account. What does AS 2301 still require?
   • Substantive procedures for each relevant assertion
   • Nothing further if controls are effective
   • Only inquiry of management
   • Removing it from the audit scope
3. Why is mapping a client's chart of accounts to a standard taxonomy important?
   • It is required by Benford's law
   • It lets models and peer comparisons treat similar accounts consistently
   • It replaces the need to test data completeness
   • It encrypts the ledger for privacy
4. Receivables grew much faster than revenue compared with peers' public filings. Which assertions does the guide link to that signal?
   • Presentation of equity
   • Completeness of payroll
   • Occurrence of lease payments
   • Existence and valuation of receivables
5. Why do scoring systems often combine rules with unsupervised anomaly detection?
   • Either approach alone tends to miss risks or flag too much
   • Rules are banned in PCAOB audits
   • Unsupervised models always need labeled fraud cases
   • Combining them removes the need for thresholds
6. What should thresholds for flagging entries connect to so teams are not flooded with alerts?
   • The number of users in the ERP
   • The client's stock price
   • Performance materiality
   • The number of staff on the engagement
7. Which check confirms that the journal entry data used for scoring is complete?
   • Counting the number of distinct users
   • Rolling entries forward from the opening trial balance to the closing trial balance
   • Running Benford's law on descriptions
   • Asking the CFO to sign the extract
8. A new product line changes how revenue should be recognized. Why might an analytics model miss this risk?
   • Models cannot read numbers
   • Revenue is never analyzed by models
   • Models only work on prior-year data from other clients
   • The model lacks business context unless people supply it