Workday is Making an AI Tool for Error and Fraud Detection in Finance

Workday signals a fundamental shift in how CFOs will approach continuous monitoring, anomaly detection, and fraud prevention as enterprise vendors race to embed AI into financial workflows.

AI Tools for Error and Fraud Detection in Finance

Finance teams have always battled duplicate invoices, misposted entries, and bad actors slipping through gaps in manual controls. But the traditional approach (periodic audits, rules-based alerts, and reconciliations) was never built for the volume and velocity of today's enterprise transactions. That approach is changing fast.

The Pleasanton, California-based HR and finance software giant is preparing a broader rollout of its Financial Test Suite. It’s an agentic AI tool built specifically to help CFOs and their teams continuously scan for fraud, errors, and anomalies in financial transactions.

"In an agentic world, the number of times you can test is almost infinite," said Tim Wakeford, Workday's vice president of financial management product strategy, in an interview. That statement alone reframes the conversation about what financial controls can look like when AI runs them in the background around the clock.

What Is the Workday Financial Test Suite?

The Financial Test Suite is an agentic AI application available on the Workday Marketplace that automates continuous financial testing across transactions, controls, and business processes. It was first unveiled in the fall of 2025 and is currently in the hands of a limited group of early adopters. General availability is expected in the second half of 2026.

The tool works by having AI agents continuously probe financial data in the background and flag items for immediate attention, investigate anomalous patterns, and, in some cases, take corrective action before errors reach the books. For example, when a duplicate invoice enters the system, the agent doesn't just flag it. It acts to stop the duplicate payment from going out.

"The software will be testing everything in the background, looking for anomalies, looking for duplicate invoices, looking for potential fraud, looking for any kind of error that exists," Wakeford noted. That continuous, autonomous scanning is the defining characteristic of agentic AI, and what separates this class of tool from legacy fraud analytics software that relies on static rule sets.

Agentic AI vs. Traditional Fraud Analytics Software

Traditional fraud analytics software has long played a role in financial controls, but it operates on predetermined rules and periodic batch runs. It catches what you've already anticipated. Agentic AI, by contrast, can reason across data, identify previously unknown patterns, and take initiative without explicit programming for each scenario.

The Capgemini Research Institute's 2025 report on agentic AI defines the distinction clearly: while traditional AI/Gen AI assistants "take actions based on user prompts and predefined logic," AI agents operate with high autonomy, "anticipate needs and take initiative," and learn and adapt continuously with long-term memory. For finance, this evolution means the system can catch fraud vectors that have never been seen before, not just those that match yesterday's rule book.

Who Else Is Building AI for Fraud Detection?

Workday isn't operating in a vacuum. The push for AI-powered financial controls has become one of the most competitive arenas in enterprise software, and a wave of vendors is racing to stake their ground.

Datarails Reimagining FP&A Infrastructure

Among the startups most aggressively targeting FP&A workflows, Datarails has made perhaps the boldest move. In March 2026, the company publicly declared that traditional FP&A software is dead and launched FinanceOS to replace it. Rather than functioning as conventional FP&A software, FinanceOS provides a governed execution layer that connects real-time, unified financial data directly to AI, enabling finance teams to leverage tools like Claude, ChatGPT, and Microsoft Copilot to build models, deploy agents, and automate workflows on trusted, secure, and fully traceable data.

The platform addresses what Datarails identifies as the core blocker to AI adoption in finance: not the intelligence layer, but the infrastructure beneath it. According to a Gartner survey cited by the company, AI adoption in corporate finance functions has essentially flatlined:

• Rising just one percentage point from 58% in 2024 to 59% in 2025

• While 91% of finance teams report low impact from their AI tools, with data quality and availability cited as the most common obstacles. 

Oracle Expands Its Finance AI Agent Portfolio

In October 2025, Oracle announced an expanded portfolio of AI agents for finance at its annual AI World conference in Las Vegas. The rollout included a new accounts payable tool, a direct parallel to some of what Workday's Financial Test Suite targets. Oracle's play is to embed AI deeply into the Oracle Fusion Cloud suite, which allows organizations already in the Oracle ecosystem to automate payables, receivables, and financial reconciliation with minimal manual intervention.

SAP and the Intelligent Finance Agenda

SAP has been building out its Business AI capabilities across its ERP platform, including in financial management. Through SAP Joule (its AI copilot) and embedded agents in SAP S/4HANA, the company is pushing toward continuous financial close processes and predictive cash flow analytics. Dr. Walter Sun, SVP and Global Head of AI at SAP, has noted that AI agents are positioned to fundamentally transform the speed and accuracy of financial operations at scale.

Salesforce’s Agentforce in the CRM Ecosystem

While primarily known for CRM, Salesforce has extended Agentforce directly into financial services workflows through a dedicated launch in May 2025. Agentforce for Financial Services integrates with Salesforce Financial Services Cloud and is designed to support insurers, banks, and wealth management companies looking to offload administrative work through agentic AI. This automates routine front-office tasks such as loan recommendations, client meeting preparation, and service requests.

More recently, Salesforce launched Agentforce Operations in April 2026, targeting back-office bottlenecks, with agents managing end-to-end processes like underwriting, extracting data from tax returns, chasing missing signatures, and validating details against compliance rules across systems.

The Economic Case of Agentic AI in Finance

The financial case for deploying AI agents in finance is compelling, and the numbers are getting harder to ignore. Based on survey responses from 1,500 senior executives across 14 countries, projected that agentic AI could generate up to $450 billion in economic value through cost savings and revenue uplift by 2028. Organizations that successfully scale AI agents in the near term are expected to generate around $382 million on average over the next three years, approximately 2.5% of annual revenue for a $15 billion company.

For finance specifically, the ROI equation is particularly sharp. Catching a fraudulent transaction or duplicate payment mid-cycle has a much higher return than flagging it post-close, and that's precisely the window that real-time AI detection opens up.

How Agentic AI Is Changing the FP&A Function

The impact of agentic AI on Finance and FP&A extends well beyond fraud detection. Across the finance function, AI agents are beginning to reshape roles, processes, and the very definition of financial control.

From Periodic Reviews to Continuous Financial Controls

Historically, financial controls operated on a schedule: monthly reconciliations, quarterly audits, and annual reviews. The problem is that fraud and errors don't operate on a schedule. Agentic AI breaks the bottleneck by making financial testing continuous.

AI Agents Handling Tasks Previously Impractical at Scale

One of the more nuanced insights is that agentic systems can unlock tasks that were previously impractical due to resource constraints, rather than just speeding up tasks that were already being done. A finance team with limited headcount might previously have sampled 5% of transactions for review. An AI agent can review 100% of them, all the time. That isn't just efficiency, but a qualitative change in the coverage and reliability of financial controls.

Rethinking ROI Metrics for AI in Finance

The shift also demands a rethinking of how CFOs measure the value of their AI investments. Traditional metrics (headcount reduction, time saved per task) may undercount the actual return. When an AI tool for error and fraud detection in finance prevents a six-figure duplicate payment or catches an early indicator of collusive fraud, the ROI isn't in the hours saved. It's in the loss avoided. That's a fundamentally different value calculus, and finance leaders who understand it early will be better positioned to make the investment case internally.

Trust, Governance, and the Limits of AI Detection

Despite the momentum, adoption of agentic AI in finance is moving carefully, and for good reason.

• The Capgemini study found that confidence in fully autonomous AI agents fell 16 percentage points over the prior year.

• With only 27% of organizations now express trust in fully autonomous systems (down from 43%).

• Ethical concerns around data privacy, algorithmic bias, and the "AI black box" were cited as widespread, but few organizations are taking decisive action to address them.

How Much Should AI Decide Alone?

Capgemini's report outlines a six-level autonomy scale for AI agents: from Level 0 (fully human-handled) to Level 5 (fully autonomous and self-evolving).

• In 12 months, only 15% of business processes are expected to operate at Level 3 or above; by 2028, that share is projected to grow to 25%.

• For finance controls, where auditability and human accountability are non-negotiable, most deployments are expected to operate in the Level 2–3 range (AI-augmented decision-making and semi-autonomous handling of complex tasks) with humans remaining actively involved in governance and exception management.

What Finance Leaders Should Watch For

For CFOs and finance technology leaders evaluating their AI strategies, a few signals are worth watching closely:

Vendor ecosystem consolidation is accelerating. Workday's $1.1 billion acquisition of Sana Labs in early 2026, which brought superintelligent agentic capabilities into its platform, illustrates how quickly the landscape is being reshaped through M&A.

Data readiness is the gating factor, but competitive pressure is also real:

• Fewer than one in five organizations report high levels of data readiness, and over 80% lack the mature AI infrastructure needed to scale agentic systems effectively.

• For AI fraud detection to deliver on its promise, the underlying financial data has to be clean, accessible, and governed, which remains a major gap for many enterprises.

• 93% of leaders believe organizations that successfully scale AI agents in the next 12 months will achieve a competitive advantage over peers. For finance functions, that advantage is increasingly tied not just to efficiency but to resilience, the ability to detect and respond to financial risk faster than any manual process allows.

For finance leaders, the question is no longer whether to embrace AI for fraud detection. It's how to govern it well enough that the benefits outweigh the risks, and how to move fast enough that the organization doesn't get left behind while the rest of the industry rewires its financial controls from the ground up.