Founder: AI companies face 'Great Sorting,' most won't survive scale
Auditoria.AI founder reflects on competitive dynamics in enterprise AI for finance, arguing that only a small class of AI companies will reach meaningful scale in the OCFO market.
The Wire — Page 3
PCAOB Chair: AI Can Only Handle Part of Auditor's Job
PCAOB chair Jim Logothetis argues AI will strengthen accounting by automating routine tasks, but cannot replace core audit judgment and client relationships.
1 more from Accounting Today
Caseware launches AI agents for Excel, upgrades client request tools
Caseware integrates AI agents into Microsoft Excel and debuts AI-first Client Requests feature, expanding automation capabilities for audit and accounting workflows.
AI agents fail consistency tests crucial for audit, study finds
Research shows AI agents return inconsistent answers 38-74% of the time on repeated tasks—a critical failure for audit and compliance. Researchers propose "skill habit formation" to improve reliabi...
8 more from arXiv: AI + Accounting
AI agents reverse-engineer business rules from legacy enterprise code
Researchers propose an agentic framework that autonomously generates Business Requirements Documents by reverse-engineering source code and configuration files, reducing manual work during system m...
Study examines whether enterprise AI agents maintain compliance under pressure
New research framework (PACT) evaluates how LLM agents handle rule violations in finance and other regulated sectors, addressing critical compliance risks as enterprises deploy AI into sensitive co...
Framework Proposes Graded Independence Standards for AI Auditors
Academic paper argues audit independence for agentic AI systems must move beyond binary classification, introducing graded standards across principal, substrate, and evidence independence dimensions.
LLM Overseers Remain Resistant to Procedural Trace Manipulation in Compliance...
Researchers test whether detailed procedural traces in LLM audit loops make judges gullible; analysis of 4,551 compliance judgments shows error detection stays near ceiling despite trace variations.
Audit Protocol Detects Supervision Drift in AI Credit-Risk Models
Researchers propose a multi-signal audit framework to detect drift in proxy-labeled credit-risk prediction models, addressing base-rate shifts and feature-label relationship changes under deployment.
Frontier AI models fail buried audit evidence despite high benchmark scores
Audit of frontier LLMs reveals shallow document-reading ability: models score well on standard benchmarks but accuracy drops sharply when evidence is buried, with increased hallucinations, tool cal...
Researchers challenge pair-counting methods in transformation audits
Academic paper argues that counting equivalent pairs overstates audit coverage in transformation systems; proposes four metrics (edge count, effective contrast rank, population support rank, spectr...
LLMs fail as automated auditors when document contamination scales, study finds
Research testing Google Gemini 3.0 Pro on 450 planted document errors reveals LLMs suffer batch-size degradation and confident hallucination—critical vulnerabilities for automated audit applications.
Havertys expands agentic AI controls testing with Optro after four-year partn...
Furniture retailer Havertys deepens its use of Optro's agentic AI platform for autonomous audit controls testing, reducing manual auditor workload and improving testing confidence.
1 more from AuditBoard Blog
CAEs must rethink controls, skills for agentic audit era
AuditBoard explores how Chief Audit Executives must adapt internal controls, guard against "agent washing," and upskill teams as AI agents reshape audit practices.
Auditing in the age of good-enough AI
Opinion piece explores how 'good enough' AI is reshaping audit practices and standards, raising questions about acceptable accuracy thresholds when AI assists or performs audit work.
Audit intake process quality determines AI tool success
Suralink argues that AI adoption in audit fails when data ingestion is poorly structured; clean, versioned intake workflows are prerequisite for effective AI deployment.
Thomson Reuters, Suralink Partner on AI-Powered Audit Workflows
Thomson Reuters integrates Suralink collaboration platform with CoCounsel Tax, Audit & Accounting to automate client engagement in audit workflows.
1 more from CPA Practice Advisor
Caseware embeds agentic AI into audit workflows
Caseware launches AI-first evolution of Client Requests tool, bringing engagement intelligence directly into audit preparation processes.
Caseware launches AI agents for audit's final workflows with Excel integration
Caseware debuts Excel and request agents targeting audit's "last mile"—the time-consuming final stages of engagement—automating manual data handling and document requests.
1 more from AccountingWeb
RSM Cayman auditors slash testing time by 90% using Excel agents
RSM Cayman external auditors deployed AI Excel agents to automate routine audit testing, reducing time spent on repetitive tasks by 90% and improving audit efficiency.