An Episode and Article in the EPM Intelligence Oracle AI World Series featuring Rich Wilkie (Senior Director, Financial Close Product Management at Oracle) and Al Marciante (Former VP, Product Management at Oracle).
Key Takeaways
- Finance is shifting to real-time decision making: AI accelerates analysis, reduces manual work, and enables finance teams to guide the business proactively rather than reactively.
- Oracle AI is secure by design: Built directly into Oracle Cloud EPM, AI features follow SOC Reporting guidelines and rigorous governance ensuring no customer data leaves the environment.
- Customer insights are shaping Oracle’s roadmap: Organizations are driving demand for deeper narrative explanations, automated variance insights, and streamlined close processes.
- Agentic AI is redefining the close: Intelligent agents orchestrate ledger, consolidation, and reconciliation tasks, reducing bottlenecks and elevating the finance function
AI’s Real Impact: Finance Evolves from Reporting to Decision-Making
A major theme from Rich and Al is how AI is transforming the role of finance itself. Finance teams used to deliver insights after the close. With AI embedded inside Oracle Cloud EPM, finance is now positioned to inform decisions continuously, not retroactively.
Rich explains that AI gives teams immediate access to insights, trends, and variance drivers, making finance a strategic partner rather than a back-office function:
“We’re looking to transform finance from a role that provides information to a role that drives decision-making.”
For Oracle customers, AI serves as a competitive advantage in terms of cost and time savings. Additionally, early adopters get the unique opportunity to influence the product development with Oracle. For these reasons, we hear many executives declare they’re “100% all-in on AI.”
How Oracle Customers Are Shaping the Next Generation of AI Features
Oracle’s roadmap is increasingly built with customers, not just for them.
Early adopters aren’t just validating AI, they’re pushing for advanced capabilities, especially in reporting and analytics:
- Narratives that go beyond visible variances
- Cross-dimensional explanations (ex: product × market × region)
- Automated commentary that reduces drilling and manual analysis
Customers want deeper insight without additional reporting overhead:
“I don’t want my users drilling and clicking… I just want the narrative, because that’s what I do as a financial analyst.”
This customer-driven innovation loop is speeding up the delivery of new features across EPM.
Security & Compliance: How Oracle Ensures AI You Can Trust
AI adoption often comes with concerns around auditability and SOC compliance. Rich emphasizes that Oracle’s AI is held to the same standard as its core SaaS applications:
- No customer data is shared or used to train models
- All AI features undergo stringent Oracle review processes
- Models operate fully within Oracle’s secure cloud boundary
“Just for us to get a feature out the door, we have to go through a really stringent review process… nothing we do is going to be shared.”
This design gives highly regulated industries the assurance they need to scale AI safely.
Why Are Financial Consolidations Still Hard?
Despite modern tools, many organizations still struggle to close the books quickly. Rich outlines the root causes:
- Multiple general ledgers from acquisitions
- Data inconsistencies across entities
- Lack of best-practice processes
- Heavy reliance on manual work
AI can dramatically reduce this friction, but process standardization still matters:
“You’re not going to solve that chaos without a significant investment in software and a driver toward better process.”
EPM Cloud provides a unified platform built around best practices, but transformation requires pairing technology with strong process design.
Moving Toward a Continuous Close: Real-Time Monitoring for Real-Time Decisions
What is a Continuous Close? An operating model where financial data is reconciled, validated, and close-ready throughout the period so period-end becomes confirmation, not chaos. This on-demand monitoring is becoming an achievable goal thanks to AI-driven automated reconciliations and streamlined data flows.
This type of “soft close” don’t require perfect accuracy. Teams can use estimates, accrual patterns, and AI surfacing anomolies early. A Continuous Close means less human bottlenecking, more real-time visibility, and better midstream adjustments.
“There’s more software. There’s more automation. There’s more AI… and that’s a big driver of how you get there.”
Agentic AI: The Future of the Close Process
One of the most exciting developments previewed at AI World is Agentic AI, autonomous agents inside Oracle Cloud EPM that orchestrate close tasks.
This is the next evolution of the close: intelligent orchestration, rather than manual checklists:
- The Ledger Agent detects activity and signals readiness
- The Consolidation Agent determines when consolidations should run
- The Reconciliation Agent identifies mismatches and resolves them
- A Supervisor Agent coordinates them all
- Meanwhile, the finance professionals intervene when judgment is needed
Rich describes it as groundbreaking:
“Having these agents manage the process is what’s going to transform finance.”
Where Customers Can Start Tomorrow
Even as agentic AI evolves, Oracle already ships close-focused AI features today:
- Job analytics summarizing where consolidation time is spent
- Natural language “script narration” to generate calc scripts
- Machine learning for transaction matching predictions
- Early adopter programs with hands-on support from Oracle PMs
These small wins help finance teams build confidence and momentum.
EPMI’s Key Insights
AI, ERP, and EPM are reshaping enterprise performance by moving intelligence directly into the systems where decisions are made. As Rich and Al emphasized, the real challenge isn’t adopting AI, it’s navigating where it can immediately streamline the close and elevate insights. In the world of financial operations, AI isn’t replacing finance teams; it’s accelerating their ability to analyze, explain, and guide the business.
At EPMI, we help organizations start with clear goals, unify their data, and empower leaders to make better business decisions using Oracle’s built-in AI and enterprise solutions.
FAQs
1. Does Oracle AI require a separate implementation?
No. Most AI features activate directly inside Oracle Cloud EPM and work with your existing data and security model.
2. Does Oracle train AI models on customer data?
No. Customer data never leaves Oracle’s secure environment and is not used to train public models.
3. What are the biggest close bottlenecks AI can help reduce?
Fragmented ledgers, inconsistent processes, data quality issues, and manual reconciliations.
4. Can AI help organizations move toward a continuous close?
Yes. AI automates reconciliations, highlights exceptions, and provides mid-period visibility.
5. What is Agentic AI and why does it matter?
Agentic AI uses autonomous agents to orchestrate tasks like ledger updates, consolidations, and reconciliations—minimizing manual touchpoints and accelerating the close.
Watch the AI World Series on YouTube!