Oracle Demonstrates Embedded AI for Productivity Gains

Oracle held an industry analyst summit recently where the focus was on artificial intelligence (AI) and embedded AI. At the event, Oracle demonstrated progress in adding useful AI-enabled capabilities to its business applications, especially in finance and accounting, supply chain, HR and revenue management. To put this into context, across the software industry, AI is already at work in many finance-focused applications that are currently available, albeit often in limited release. We are in the early stages of a new generation of business applications. After decades of limited innovation, software providers are in a race to add AI capabilities to their software as fully and rapidly as possible without frustrating users, while retaining a reputation for reliability and security. I recently wrote about the ability of AI and generative AI (GenAI) to boost productivity and the need for guardrails and risk management in applying this technology. Buyers and users of business applications should expect a steady stream of announcements from software providers for the foreseeable future.

In the race to innovate, larger, incumbent providers have an advantage. This goes against the standard model, up to now, of software industry disruption where, as new technology is introduced, smaller, nimbler companies are able to exploit faster than “legacy” providers. In the past, the new players were able to reconceive how new technology could accomplish business tasks, unburdened by legacy code and incompatible sales or go-to-market models. In this meme, the old guard are dinosaurs, ready for extinction.

However, that’s not the case with AI in business software applications because the technology is additive to existing products’ capabilities. In this case, the larger incumbents, including Oracle, have four key advantages. They have:

• Broad and deep functionality built into mature, stable applications.
• Large data sets from diverse groups of users on which to train systems anonymously.
• A large installed base on which to spread fixed development costs.
• Scale to negotiate better input factor pricing such as memory and compute power.

These advantages allow the large, incumbent providers to include almost all core AI capabilities at no cost to customers, potentially giving them an advantage over smaller or less well-established competitors that either must find ways to increase revenues, be more selective in developing and embedding AI in their software or deal with lower profitability. In Oracle’s case, managing its own Cloud Infrastructure can provide advantages to some customers or potential customers.

ISG-Ventana Research asserts that by 2027, almost all providers of software designed for finance organizations will incorporate AI capabilities to reduce workloads. The productivity gains from AI are more likely to be achieved initially through a steady stream of small hacks and initiatives that eliminate less productive and unproductive work, rather than big bangs. One reason is that the process of infusing AI into cloud business applications will be incremental, with the level of sophistication constrained by a gradual leveling up of data availability and proven trustworthiness.

One reason why so much embedded AI is now available is that for Oracle, like many established business software providers, this is not new. Work in this area has been underway for more than a decade, but the attention paid to GenAI catalyzed market demand and spurred product innovation. The sessions at the analyst event highlighted several areas where Oracle has AI-enabled offerings.

ERP and other transactions management systems are ripe for AI innovation. Technology can enable software users to achieve productivity gains through automation and end-to-end data quality management using predictive and GenAI. For example, Oracle demonstrated “touchless” operations where incoming business documents are machine read using optical character recognition and the relevant data is extracted with no human intervention. Moreover, the data captured can be programmatically enriched for context to (for example) properly classify amounts for cost allocations, automatically match an invoice to a purchase order or some other clear action. Transactions can be immediately auto-reconciled, reducing period-end tasks and supporting a continuous accounting approach to financial management. Eliminating humans from repetitive tasks increases productivity directly. A touchless approach also can significantly reduce the incidence of human error (such as fat-fingering amounts) that drive additional downstream work devoted to finding and correcting mistakes. Using relatively simple machine learning (ML) can spot potential errors or missing information better than a fixed validity check, while predictive and GenAI can quickly offer individuals options for resolving an anomalous situation. Increased productivity can reduce costs and boost buyer and supplier satisfaction. Eliminating dull repetitive work while significantly reducing workload spikes can make it easier to attract and retain scarce departmental talent as the staff can maintain a satisfying work/life balance. ISG-Ventana Research asserts that by 2027, almost all procure-to-pay software suites will use AI and optical character recognition OCR to automate data ingestion from external documents and emails, saving time and ensuring all necessary information is captured accurately at the source.

Oracle also pointed to the ability of touchless operations in the near future to support what it calls a continuous close, which is an element of continuous accounting. (From an accountant’s perspective, this is not a true close but a close-enough close.) Using AI enables the greater use of straight-through processing supported by, for example, automated matching, auto-resolution of exceptions and automated classification to handle an increasing percentage of the accounting workload. Even in handling the remaining exceptions, GenAI can present sets of recommendations to improve the productivity of a skilled human in resolving those exceptions.

Predictive analytics have been around for decades but have not been widely adopted in finance departments because of data breadth, data quality and model-building skills issues. As enterprises seek to make broad use of AI technologies, they now have a substantial set of more capable tools at their disposal to ensure they have the accurate and accessible data they need. For example, as my colleague Matt Aslett explains, data orchestration technologies enable and facilitate the flow of data across the organization to support the practical uses of AI. Predictive analytics supported by ML facilitates the creation and maintenance of models for forecasting, planning and budgeting. Oracle has the infrastructure to support enterprise data management requirements and would do well to facilitate the bidirectional movement of data to and from Oracle and non-Oracle applications to support enterprise forecasting and planning.

Oracle illustrated several available predictive AI use cases. For example, the CFO, treasury and financial planning and analysis organizations need to have an accurate (as possible) view on cash collections. Based on expected sales, individual customer or paying-agent payment histories, collection efforts and potentially other factors that historically have had a meaningful impact on incoming payments, Oracle’s system will forecast collections over the desired period presented as expected, best- and worst-case scenarios. Because it’s regression analysis, it’s possible to inform the user of the potential accuracy of the forecast as well as the most important factors in the regression analysis that cause results. This sort of explainability is essential to user acceptance and intelligent use of these forecasts.

Reducing frictions in payment processes is a growing priority in finance departments because this permits better control of cash while enhancing productivity. At the 2022 CloudWorld, Oracle announced its B2B Commerce platform that provides integration with J.P. Morgan for treasury services, trade, commercial card and merchant services capabilities for integrated banking as well as travel card and expenses services. Oracle Fusion ERP has payment technology embedded to enable invoice approval automation as well as the use of virtual cards for supplier payments. Technology can eliminate sources of delay in processing invoices, making it possible for a buyer to negotiate and execute early payment discounts with suppliers, thereby enhancing profitability. Buyers with stronger credit than suppliers can also employ reverse factoring. This is a method of supply chain finance that optimizes working capital utilization by having a financial intermediary provide reliable or even early payment to suppliers while enabling buyers to defer their payment to that intermediary.

On the seller side, Fusion ERP and NetSuite connect accounts receivable management with cash collection via ACH and virtual cards. Payment technology is vital to accounts receivable automation, which has become increasingly important as business-to-business selling has grown more complex. Buyers want a simpler and more streamlined experience similar to what they expect as individual consumers, but it takes technology to make the complexities of today’s B2B selling appear simple. Processing transactions is more difficult than ever because the structure of these transactions can be complex. While one-time sales are straightforward, B2B purchases may be covered by a negotiated annual contract that sets pricing, discounts, terms and conditions. Connecting the payments and receivables processes is necessary to make that experience feasible.

Enterprises are already making significant investments in AI. ISG Research finds that, on average, organizations spent 2% of IT budgets on AI in 2023. The average expected spend for 2024 is 3.7% and 5.9% in 2025. The research also finds that enterprises are willing to spend more per seat to have AI-enabled capabilities, with sales performance management, supply chain management and treasury and risk management showing the greatest propensity to pay more.

Typically, finance and accounting departments have proven to be technology laggards in adopting new methods. An innate conservatism, aversion to risk and the need to ensure complete accuracy are the human factors at work. However, the benefits of AI-enabled applications to productivity and staff morale are compelling. I strongly recommend that finance executives adopt a fast-follower approach to technology adoption. A fast-follower approach is now a necessity because software designed for finance and accounting departments is evolving rapidly, and incremental adoption as AI capabilities become available is a more productive, less disruptive and less risky approach than playing catch-up. AI-enabled applications will reduce a significant part of the department’s workload currently spent on repetitive tasks and mechanical processes, allowing staff to focus on the more valuable work that requires expertise, experience and judgement.

Regards,

Robert Kugel