Workday Shifts Financial Controls From Audits to Real-Time Testing
Workday is embedding autonomous AI agents directly into financial transaction flows, moving financial controls from periodic review cycles to continuous, real-time monitoring. The company's Financial Test Suite, in limited release now and expected to reach general availability in the second half of 2026, runs in the background to detect fraud, errors, and anomalies as transactions occur.
The system can identify and stop issues-such as duplicate invoices-before payments are processed. This represents a fundamental shift in how financial controls operate inside ERP systems.
From Month-End Reviews to Continuous Testing
Finance teams have traditionally relied on sample-based audits and month-end reviews to catch problems. The Financial Test Suite changes that model by testing transactions continuously as they flow through the system.
"In an agentic world, the number of times you can test is almost infinite," Tim Wakeford, Workday's VP of financial management product strategy, said in an interview with CFO Dive.
Control logic shifts from something finance teams design around audit cycles to something embedded directly into transaction processing. That changes who owns control design-it becomes a product and architecture question, not just a policy and process one.
New Operational Risks Emerge
Stopping duplicate payments before they clear has clear value. But pre-transaction intervention also introduces new risks that audits alone cannot manage.
When an autonomous system makes real-time decisions that affect cash flow and vendor relationships, false positives and missed edge cases become operational risks. The reliability of the model matters as much as the logic of the control.
Finance leaders are showing genuine interest in agentic AI for financial workflows. But skepticism remains around validation, governance, and how these systems will be audited and overseen.
Auditability Becomes Decision Traceability
Traditional audit models rely on sampling and documentation after the fact. Continuous testing requires a different approach: explaining why a system acted, not just proving a control exists.
Autonomous systems that intervene in financial transactions need clear audit trails, override mechanisms, and control frameworks-particularly for public companies and regulated industries. Adoption will depend on whether vendors can provide transparency into agent decisions: what triggered an action, what data was used, and how outcomes can be reviewed or reversed.
Workday's push into agentic finance reflects a broader industry shift. Oracle has expanded its own portfolio of AI agents for finance, while other vendors are moving closer to enterprise workflows. The line between system-of-record platforms and AI-driven execution layers is blurring.
For finance professionals, this means the skills required to manage financial controls are changing. Understanding how autonomous systems make decisions, validating their outputs, and designing controls that work with-not against-AI agents are becoming core competencies. See AI Learning Path for CFOs for guidance on governance and implementation challenges.
Your membership also unlocks:
