{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp09_archetype_breadth150_20260804","research_id":"eoa_inverse_innovation_exp09_light_prior_art_20260804","cell_id":"computability_boundary_mapping__accounting_auditing","search_lanes":{"direct_problem_and_intervention":{"queries":["audit journal entry automated controls custom scripts static analysis timeout inconclusive","audit analytics journal entry testing automated tool false positives continuous auditing research"],"source_ids":["SRC1","SRC4"],"no_result_note":"No direct public source was found describing an audit duplicate-recognition analyzer for unrestricted client posting programs that couples a formally proved decidable fragment with explicit non-Boolean fallback routing."},"synonyms_and_historical_terms":{"queries":["software verification unrestricted programs undecidable Rice theorem halting problem static analysis sound complete termination","static analyzer unknown timeout bounded model checking three valued result sound complete decidable fragment"],"source_ids":["SRC2","SRC3"],"no_result_note":null},"products_practices_and_standards":{"queries":["official standard audit automated application controls journal entries ISA 315","site:sv-comp.sosy-lab.org competition rules UNKNOWN timeout verification results"],"source_ids":["SRC1","SRC3"],"no_result_note":null},"component_combination":{"queries":["decidable fragment program verification unknown result three valued static analysis primary research","audit analytics journal entry testing automated tool false positives continuous auditing research"],"source_ids":["SRC2","SRC3","SRC4"],"no_result_note":null}},"sources":[{"source_id":"SRC1","title":"ASA 315: Identifying and Assessing the Risks of Material Misstatement","publisher":"Auditing and Assurance Standards Board (Australia)","url":"https://standards.auasb.gov.au/asa-315-mar-2021","source_type":"OFFICIAL_STANDARD","claims_supported":["Controls over journal entries, automated or manual, are expected to be identified in all audits.","Auditors must evaluate automated controls and risks arising from IT, obtain sufficient appropriate evidence, document significant judgments, and revise risk assessments when inconsistent new information appears.","Automated techniques may assist journal-entry analysis but do not displace the auditor's responsibility for risk assessment and further procedures."]},{"source_id":"SRC2","title":"Decidable Verification of Uninterpreted Programs","publisher":"Proceedings of the ACM on Programming Languages","url":"https://arxiv.org/abs/1811.00192","source_type":"PRIMARY_RESEARCH","claims_supported":["Automatic verification of the studied unrestricted program class is undecidable in general.","A formally restricted coherent-program subclass admits decidable verification in PSPACE.","The paper exemplifies moving from an undecidable general verification problem to an explicitly characterized decidable program fragment."]},{"source_id":"SRC3","title":"SV-COMP 2026 Competition Rules","publisher":"International Competition on Software Verification","url":"https://sv-comp.sosy-lab.org/2026/rules.php","source_type":"OFFICIAL_GUIDANCE","claims_supported":["Verification runs that reach the time limit are terminated and recorded as timeout.","A timeout is interpreted as UNKNOWN rather than TRUE or FALSE.","The rules operationalize separation between correct verification results and resource-bounded non-results."]},{"source_id":"SRC4","title":"Anomaly Detection in General Ledger Data: Results from a Hybrid Approach","publisher":"German Research Center for Artificial Intelligence (DFKI)","url":"https://www.dfki.de/en/web/research/projects-and-publications/publication/15550","source_type":"PRIMARY_RESEARCH","claims_supported":["Journal Entry Tests generate large result lists requiring substantial additional auditor effort.","False positives are a recognized practical problem in automated general-ledger analysis.","The reported hybrid anomaly-detection experiments used synthetic journal-entry data."]}],"problem_evidence":{"status":"PARTLY_SUPPORTED","finding":"The component problem is visible: an official audit standard treats journal-entry and automated-IT controls as audit-relevant, audit research reports burdensome false positives, and software-verification practice explicitly preserves timeout as UNKNOWN. The bounded search did not establish that audit firms currently submit unrestricted posting scripts to a universal duplicate-recognition analyzer or coerce its timeouts into clean or failed Boolean control conclusions.","source_ids":["SRC1","SRC3","SRC4"]},"closest_prior_art":[{"name":"SV-COMP explicit UNKNOWN and timeout result semantics","source_ids":["SRC3"],"overlap":"Directly overlaps the proposal's rule that timeout or analyzer failure must not be interpreted as proof that a property holds or fails.","remaining_difference":"It is a verification-competition protocol, not an accounting-assurance contract; it does not define economic-event identity, an admitted posting-language fragment, engagement-partner disposition, or audit reclassification triggers."},{"name":"Coherent-program decidable verification boundary","source_ids":["SRC2"],"overlap":"Closely overlaps formal specification of a verification problem, proof of general undecidability, and identification of a restricted program subclass permitting total decision.","remaining_difference":"The verified property and computation model differ, and the work does not combine fragment membership, bounded evidence routing, accounting assertions, or audit authority controls."},{"name":"Automated journal-entry control assessment and hybrid anomaly analysis","source_ids":["SRC1","SRC4"],"overlap":"Establishes the accounting setting, automated journal-entry analysis, IT-control evaluation, false-positive burden, synthetic testing, and continued auditor judgment.","remaining_difference":"These sources do not classify duplicate-recognition as a computability problem, publish soundness/completeness/termination guarantees, or preserve proof, bounded non-observation, unknown, unsupported input, and service failure as separate assurance states."}],"prior_art_disposition":"ADJACENT_PRIOR_ART","contrastive_claim_remaining":"For client-defined posting programs, a versioned audit-tool contract can combine a checked undecidability classification for the unrestricted duplicate-recognition question, an enforceable decidable-language fragment, and an audit-governed router that prevents bounded non-observation, timeout, unsupported input, or failure from becoming proved absence. The searched sources support each major ingredient separately but did not disclose this accounting-specific combination.","contrastive_claim_falsifier":"The claim would be falsified by finding an existing accounting-assurance method that already combines a formally checked unrestricted-program impossibility result, a conservatively enforceable posting-language fragment with a total exact duplicate-recognition procedure, explicit unknown/out-of-fragment/failure outputs, version-triggered reclassification, and retained engagement-partner authority; it would also fail operationally if the proposed fragment or event identity cannot be checked soundly or users continue treating fallback results as proved absence.","gates":{"adequate_source_search":{"status":"PASS","rationale":"The bounded search covered the proposal directly, computability and verification terminology, audit standards and journal-entry practices, decidable subclasses, timeout semantics, and combinations of those components. Four opened sources span four publishers and include official and primary sources.","source_ids":["SRC1","SRC2","SRC3","SRC4"]},"supported_problem":{"status":"PASS","rationale":"The exact alleged audit-tool behavior was not documented, but official and primary evidence supports the materiality of automated journal-entry controls, false-positive/manual-review burdens, and the need to distinguish timeout from a decided result; therefore the problem is partly supported.","source_ids":["SRC1","SRC3","SRC4"]},"distinct_testable_claim":{"status":"PASS","rationale":"No retained source disclosed the full accounting-specific boundary contract. Its remaining claim is falsifiable through a checked reduction, fragment-membership and termination proofs, output-routing tests, and a search for an existing method containing the same combination.","source_ids":["SRC1","SRC2","SRC3"]},"bounded_next_test":{"status":"PASS","rationale":"A methods sprint using no more than 20 synthetic programs can test the reduction direction, conservative fragment membership, analyzer termination, five-state output behavior, and accounting fidelity without deployment or effect-size claims.","source_ids":["SRC2","SRC3","SRC4"]},"no_obvious_safety_or_authority_stop":{"status":"PASS","rationale":"A read-only study on synthetic or authorized de-identified programs presents no obvious safety stop if it cannot alter ledgers or audit conclusions. The official standard supports retaining auditor judgment, evidence evaluation, documentation, and further-procedure authority.","source_ids":["SRC1","SRC4"]}},"screen_survival":true,"world_novelty_boundary":"This four-source bounded public-web screen establishes only that the proposal is researchable and adjacent to known audit analytics, decidable verification subclasses, and explicit UNKNOWN verification semantics. It cannot establish world novelty, patentability, market size, expert acceptance, implementation feasibility across actual client languages, or realized audit value."}