With over 40 years of leadership in the enterprise applications field, many customers run their businesses with Oracle Applications Unlimited Family, a competitor such as JD Edwards, PeopleSoft, Siebel, Oracle E-Business Suite, or SAP. However, with the recent explosion of AI, many business leaders are trying to understand how to leverage AI within legacy applications. Earlier this year, Forbes reported that six of the 10 large companies around the world were using generated AI. Of that group, 74% saw a significant return on investment, while 45% doubled employee productivity.1 AI provides real-world value and enables those who use it to travel faster and more efficiently. These productivity gains are equalizers for small organizations and help them scale efficiently to disrupt their competitors and markets. However, as AI adoption grows and these types of efficiency become more common, those waiting to adopt AI could unfortunately fall behind their competitors soon.
Choosing a path to AI adoption
Leaders are responsible for the board to adopt an AI strategy that will allow them to compete in today's AI-enabled market, and need to move quickly to respond to the market and avoid disruption. Oracle's Applications Unlimited program will continue to be supported until 2035, but leaders will need to determine whether it makes sense to add AI to their legacy environments or whether they need to migrate to an updated platform that already has AI built in. For example, Oracle Fusion Cloud applications offer a complete suite of modern best practices across Finance, HR, Supply Chain, and CX, with over 150 AI capabilities. Additionally, the rapid pace of innovation in Fusion Applications and quarterly update cycles allow leaders to maximize their investments while running AI strategies within their core business processes.
Migrating from legacy platforms to Oracle Fusion applications may seem too difficult or complicated for an organization with a well-established, highly customized system, but building effective AI tools is much more difficult. Despite all the potential benefits of AI, many companies struggle to implement their AI vision. A recent Wall Street Journal article found that while many executives are optimistic about AI, AI is far more work than initially expected.2 There are many AI services and tools on the market. However, to implement an effective AI strategy, data science skills, development expertise, specialized infrastructure, deep business process knowledge, and high quality, reliable data. While most companies may have some of these features, most companies do not have all the features they need to implement a successful AI strategy that delivers value.
General AI challenges
As leaders dive deep into their AI journey, they face many common challenges. Executives with multiple different systems or highly customized on-premises applications face a difficult battle to clean up data to begin their AI journey. Assuming that data can be aggregated, training a model on historical data may seem simple at first, but business process changes, exceptions, and outliers in the dataset can all affect AI accuracy. These factors can lead to model drift as the tool is used. Furthermore, AI expertise is in high demand and it is difficult to find, attract and maintain AI talent. A deep understanding of the business processes required to make an AI project successful means that you need buy-in from business owners who may have a different vision for how AI tools work.
Graphics Processing Units (GPUs) that run AI models are affected by supply chain shortages. GPUs need a large amount of electricity that can impact sustainability goals. Additionally, running multiple GPUs in tandem to perform complex tasks requires a lossless network. All of these factors can lead to extremely high costs of implementing AI solutions.
In addition to the technical challenges of AI implementation, recent regulatory initiatives such as the EU AI Act3 Canadian Artificial Intelligence and Data Law4and existing privacy laws such as the California Consumer Privacy Act.5 General Data Protection Regulations in the EU6when AI is implemented in a non-integrative way, everything poses a risk of regulation. The use of AI in employment practices could also pose increased risk and scrutiny from regulators.7 As a result, IT executives face the challenge of implementing their company's AI strategy and delivering value amid a rapidly changing landscape of technology, talent and regulations.
Building a successful AI strategy
A successful AI strategy cannot focus solely on technology solutions. To get the most value, effective strategies need to be multifaceted. A study published by the National Economic Research Agency shows that the greatest productivity gains when using AI will require complementary processes, organizational restructuring and adaptability development as individuals use new AI tools.8
Effective AI strategies require new ways of thinking, such as building AI into core business processes instead of bolting it to existing processes. While Genai tools can certainly provide value, organizations with highly customized legacy applications can struggle to realize the benefits of transformation by applying AI tools to undesigned processes with original focus in mind.
It is also important that executives focus on data privacy and security when selecting embedded AI solutions. Some AI vendors contractually require users to allow anonymized data mining so that vendors can monetize their company's data and intellectual property. However, due to data differences, aggregating multiple data sets produces only minimal benefits.
Furthermore, AI strategies require that humans be included in the loop of AI processes. This includes giving humans the ability to intentionally turn on AI tools, accept and reject recommendations. The NIST AI Risk Management Framework identifies that having humans in the loop of AI processes helps to promote fair and equitable outcomes.9
Finally, AI strategies need flexible and future evidence. AI innovation moves much faster than the adoption of other previous transformational technologies, such as the Internet and smartphones. Two years ago, the generated AI exploded into the scene, democratizing AI for the masses, allowing natural language interactions with data. A year later, tools such as the searched Higher Generation (RAG) allowed AI to provide context-related answers without the need to train custom models. Today, Agent AI, a new type of AI, is committed to combining the powers of machine learning, advanced analytics, and context-generating AI to automate processes and provide enhanced business value. With multiple AI technologies and disciplines, Agent AI is a challenging effort for everyone except the most mature and sophisticated organizations. Unfortunately, due to this complexity, CIO Magazine reports that an estimated 75% of companies trying to build their own AI agents fail.10
Therefore, the rapid innovation and update cycle of Software-as-a-Service (SAAS) technology requires executives to consider SAAS-first strategies using embedded AI to achieve rapid and meaningful increased AI productivity. In short, due to the complexity of AI, don't build anything you can buy.
How Oracle can help
Within AI providers, Oracle is uniquely positioned to help executives implement effective AI strategies using built-in AI technology. By embedding a wide range of predictive and generative AI and AI agent capabilities directly into everyday process flows, Oracle continuously expands the capabilities of fusion applications, democratizing AI, ensuring that all end users benefit without becoming data scientists. Oracle AI Agent Studio for Fusion Applications is a design time environment that provides the tools that allow customers to create, customize, validate, and deploy Genai features and AI agents to meet their specific needs. Unlike highly customized legacy applications, Fusion Applications' integrated data model provides the consistent, high quality data needed for effective AI use. Additionally, Oracle Cloud Infrastructure (OCI) computing power for running AI is included in Oracle Fusion applications at no additional cost.
With the CIA as your first customer, security is in Oracle's DNA. As a result, Oracle Fusion Applications AI does not allow customer data to be shared with Oracle or third-party LLM providers. With AI tools loops and quarterly release cycle loops, Oracle Fusion Applications' AI is human-centric, future proof, ready to be implemented immediately.
Because the rapid pace of innovation is combined with the risk of lagging, executives should look to Oracle Fusion applications to implement their AI strategies. Oracle enables organizations to adopt modern best practices, leverage AI-powered analytics, and use agent AI out of the box. Oracle Fusion applications using embedded AI allow organizations to focus their AI efforts on organization-specific use cases. And rather than building AI tools and bolting them to legacy applications, Oracle Fusion Applications provides leaders with a path to recognize the value of AI across their organization.
Next Steps
If your organization is ready to take the next step using AI built into your Oracle Fusion applications, contact us any time through our Oracle Cloud Application Sales Team.
For more information about AI in Fusion applications, see Oracle AI for Fusion applications. Also, to find additional information about the underlying technology powered Oracle Cloud applications, start with an introduction and check out our blog post series.
The author is a member of Oracle's North American Application Office of Technology and Innovation, and is dedicated to helping customers modernize their business through innovation. This team provides expertise and vision on subjects in AI, SaaS, platform technology, operations, and data management.
