Predictika · journal-entry audit: fraud-risk patterns, sign-off order, escalation and evidence
An AI audit agent reviews journal entries for a fictional client and proposes an audit action on each one: test it, clear it, approve it or conclude it appropriate. Before the action is recorded, the agent sends it to Predictika’s validator, which checks it against audit-methodology rules and returns every finding with the standard behind it.
This validator checks the action an AI audit agent proposes on a journal entry (testing it, clearing it, approving it or concluding it appropriate) against the rules below. It handles any entry, preparer, acting identity and engagement in its data, not just the scenarios shown here.
Any proposed audit action that touches these rules is in scope — not only the scenarios pictured here. Not covered: the underlying transactions or the financial statements; the validator produces no price or amount of its own, and it is not a substitute for an auditor’s judgment or the firm’s own methodology. The clients, preparers and entries are fictional and the ledger data is synthetic.
9 scenarios in 4 groups · each group runs from the agent’s first attempt to the outcome · click one to switch
Proposing agent: AI sales agent (scripted)
Before sending, the agent must call Recorded run
What this means
Our assessment of how typical entry-by-entry sampling and rule scripts handle the same cases. It is a judgment about effort and reliability, not a claim that nothing can be built.
| What the validator caught | Typical sampling or rule-script approach | Predictika |
|---|---|---|
| Entries that are small alone but large together | Only with workarounds Sampling and rule scripts test one entry at a time. Summing a preparer's other entries on the same engagement needs a separate analytic step that rarely runs at the moment an action is proposed. | The preparer's other entries on the engagement are part of the check, made on every call. |
| A reversal posted right after period close | Only with care Linking a reversal to the entry it reverses, and to the dates of both, means a custom report or a rule per period. | The reversal names the entry it reverses, and the dates are compared in the same pass. |
| Who may sign, and in what order | Only with care Senior review, manager approval and partner sign-off in order, never by the preparer, usually live in a workflow tool rather than in the check on the conclusion. | Sign-off order, preparer independence and the escalation thresholds are checked together. |
| One wrong identity and client that break several rules | Only with care Independence, engagement scope, evidence, documentation and escalation each need their own rule, and a person fixing the identity meets the remaining ones one at a time. | The acting identity, the client and the entry are in one request; every rule that reads them is checked together on every call. |
| Evidence behind a conclusion | Only with workarounds A conclusion statement is free text; nothing ties it to a retrievable document unless someone checks. | A conclusion cannot be recorded as appropriate without a linked, registered document. |
Every rule is checked on every call, in any order the choices arrive; one problem never hides another. Blocking findings stop the audit conclusion; approval findings need a person to sign off; everything else is computed.
| 1 | This entry, combined with this preparer's other entries on this engagement (the total held in the entry's data record), crosses performance materiality even though this entry alone does not. Standard: AU-C 240.32(a)(i); AS 2401.61 (aggregation of individually immaterial misstatements). |
| 2 | Journal entry meets at least one fraud-risk selection criterion and cannot be cleared without testing. Standard: AS 2401.61-.62; AU-C 240.32(a)(ii). |
| 3 | This entry reverses another entry posted right before period close, with the reversal itself posted shortly after close and no documented business justification on file - a classic income-smoothing pattern. Standard: AS 2401.61 (post-closing entries with little or no explanation). |
| 4 | Entries posted to an unusual account, a SOX-scoped account, or prepared by an identity with a segregation-of-duties conflict require partner sign-off, not manager approval alone, before conclusion. Standard: AS 2401.58; AS 1201 (supervision scaled to risk). |
| 5 | Entry amount exceeds 3x performance materiality and requires partner sign-off regardless of any other risk flag. Standard: Firm engagement quality control policy - dollar-threshold-scaled review (AS 1201 supervision principle; not a specific numbered provision). |
| 6 | This account is designated for system-generated postings only, but this entry was posted manually. Standard: AS 2401.61 (entries made by individuals who typically do not make journal entries, to unrelated, unusual, or seldom-used accounts). |
| 7 | The preparer cannot also review or approve their own test. Standard: AS 1201 (Supervision of the Audit Engagement). |
| 8 | Proposed action targets a journal entry outside this engagement. Standard: Firm QC policy - engagement scope (QC 1000 / ISQM 1). |
| 9 | Acting identity has an advisory-side conflict with this audit client. Standard: SEC Reg S-X, Rule 2-01 (auditor independence). |
| 10 | Cannot conclude a journal entry appropriate with no linked, traceable evidence. Standard: AS 1105 (Audit Evidence). |
| 11 | Manager approval requires a completed senior review first. Standard: AS 1201 (Supervision of the Audit Engagement). |
| 12 | Partner sign-off requires a completed manager approval first. Standard: AS 1201 (Supervision of the Audit Engagement). |
| 13 | This entry's sample selection is locked and cannot be changed. Standard: AS 2315 (Audit Sampling). |
| 14 | No documented conclusion statement in the workpaper. Standard: AS 1215 (Audit Documentation). |
| 15 | This entry combined with a related entry crosses performance materiality even though neither does alone; it cannot be cleared or concluded without testing the combined pattern. Standard: AU-C 240.32(a)(i); AS 2401.61 (aggregation of individually immaterial misstatements). |
| 16 | This entry was posted into an already-closed period with no documented override justification. Standard: AS 2401.61 (post-closing entries with little or no explanation). |
| 17 | This entry has no formal general-ledger account number attached, a pattern PCAOB's own journal-entry guidance names for entries made before or during financial-statement preparation. Standard: PCAOB Audit Focus - Journal Entries (staff publication) - entries lacking account numbers, made during financial-statement preparation; same AS 2401.61 fraud-risk-factor family, exact paragraph not independently verified. |
| 18 | Intercompany entries feed consolidation eliminations and are a named PCAOB red-flag pattern; this one cannot be concluded without partner sign-off. Standard: PCAOB Audit Focus - Journal Entries (staff publication) - accounts containing intercompany transactions; same AS 2401.61 fraud-risk-factor family, exact paragraph not independently verified. |
| 19 | This account has a known history of errors or has not been reconciled on a timely basis, a named PCAOB red-flag pattern; this entry cannot be concluded without partner sign-off. Standard: PCAOB Audit Focus - Journal Entries (staff publication) - accounts prone to errors or not reconciled on a timely basis; same AS 2401.61 fraud-risk-factor family, exact paragraph not independently verified. |
| 20 | This entry's amount shows a suspicious digit pattern (e.g. a consistent ending number across this preparer's entries), a named PCAOB red-flag pattern distinct from a plain round dollar amount; it cannot be cleared without testing. Standard: PCAOB Audit Focus - Journal Entries (staff publication) - round numbers or a consistent ending number; same AS 2401.61 fraud-risk-factor family, exact paragraph not independently verified. |
The recorded result of the validator for each scenario on this page: the request the AI agent sent, the verdict, and every finding, in the order they appear above.
The agent's request. I am the advisory manager 02, working under the Northwind Logistics engagement. For the Solara Medical Instruments inventory Reserve entry of USD 15,468.25, I propose to conclude the entry as appropriate. The entry is journal entry je_0022, prepared by agent_prep_01; there is no linked evidence. It is ready to record.
Verdict: Blocked.
The agent's request. I am the senior reviewer 02, working under the Solara Medical Instruments engagement. For the Solara Medical Instruments inventory Reserve entry of USD 15,468.25, I propose to conclude the entry as appropriate. The entry is journal entry je_0022, prepared by agent_prep_01; there is no linked evidence. It is ready to record.
Verdict: Blocked.
The agent's request. I am the senior reviewer 02, working under the Solara Medical Instruments engagement. For the Solara Medical Instruments inventory Reserve entry of USD 15,468.25, I propose to select the entry for testing. The entry is journal entry je_0022, prepared by agent_prep_01; there is no linked evidence. It is ready to record.
Verdict: Allowed.
The agent's request. I am the senior reviewer 02, working under the Northfield Foods (fictional) engagement. For the Northfield Foods (fictional) COGS - supplier discounts entry of USD 38,500, I propose to conclude the entry as appropriate. The entry is journal entry je_nf_0412, prepared by agent_nf_proc_07; it is supported by Supplier 4471 email confirming Q1 volume rebate; a senior review is on file. It is ready to record.
Verdict: Blocked.
The agent's request. I am the senior reviewer 02, working under the Northfield Foods (fictional) engagement. For the Northfield Foods (fictional) COGS - supplier discounts entry of USD 38,500, I propose to select the entry for testing. The entry is journal entry je_nf_0412, prepared by agent_nf_proc_07; it is supported by Supplier 4471 email confirming Q1 volume rebate; a senior review is on file. It is ready to record.
Verdict: Allowed.
The agent's request. I am the senior reviewer 02, working under the Northfield Foods (fictional) engagement. For the Northfield Foods (fictional) product revenue entry of USD 96,000, I propose to clear the entry as not selected for testing. The entry is journal entry je_nf_0590, prepared by agent_nf_rev_02; there is no linked evidence; it reverses je_nf_0561. It is ready to record.
Verdict: Blocked.
The agent's request. I am the senior reviewer 02, working under the Northfield Foods (fictional) engagement. For the Northfield Foods (fictional) product revenue entry of USD 96,000, I propose to select the entry for testing. The entry is journal entry je_nf_0590, prepared by agent_nf_rev_02; there is no linked evidence; it reverses je_nf_0561. It is ready to record.
Verdict: Allowed.
The agent's request. I am the senior reviewer 02, working under the Northfield Foods (fictional) engagement. For the Northfield Foods (fictional) contract revenue - SOX scoped entry of USD 172,000, I propose to conclude the entry as appropriate. The entry is journal entry je_nf_0802, prepared by agent_nf_con_05; it is supported by Signed customer contract 5530, Q2 2026; a senior review is on file; a manager approval is on file. It is ready to record.
Verdict: Blocked.
The agent's request. I am the partner 01, working under the Northfield Foods (fictional) engagement. For the Northfield Foods (fictional) contract revenue - SOX scoped entry of USD 172,000, I propose to record my partner approval. The entry is journal entry je_nf_0802, prepared by agent_nf_con_05; it is supported by Signed customer contract 5530, Q2 2026; a senior review is on file; a manager approval is on file. It is ready to record.
Verdict: Allowed.
The text of the narrated walkthrough, step by step.