08-07-2025
When The Old Becomes New: Revisiting Opportunities In Abandoned AI Use Cases
Beena Ammanath - Global Deloitte AI Institute Leader, Founder of Humans For AI and Author of "Trustworthy AI" and "Zero Latency Leadership."
Organizations are searching for valuable AI use cases, but the most transformational opportunities might not be new. Instead, some of the best ideas may be found in the pile of past technology pilots that did not make it to production.
Generative AI, and more recently, agentic systems, are maturing quickly. With GenAI, early applications have been deployed for customer-facing and back-office activities, such as chatbots for call centers or AI-generated summaries of legal documents. Yet, commoditized GenAI applications such as these are available to every enterprise at this point. The bolder value vision is in using new AI capabilities to solve long-standing inefficiencies or problems that may have been targeted before, albeit with inferior technology.
In the hunt for use cases, business leaders should reexamine projects that were previously attempted but abandoned because the necessary technology was nascent or did not yet exist.
Looking Back To Point The Way Forward
Here's an example of how an idea conceived years ago may be ripe for today's technology landscape. A global manufacturer maintains 400+ ERP systems, and the number of systems complicates procurement and leads to significant discrepancies in product costs across business units, even when parts are ordered from the same vendor. What's more, the procurement leader has limited visibility into what different groups in the company are paying for parts, the data is fragmented and unstructured and some data exists only on paper. The opportunity is to improve visibility into data and ERP systems to optimize costs.
Even a decade ago, technology was not ready for this use case. I know because I led a team that attempted to build this kind of solution. We pulled prices and item descriptions from the central data lake and built an application to give procurement leaders more visibility into parts data. The significant savings generated inspired us to reach farther with automation.
One issue was that parts descriptions used technical terms and numbers that were difficult to decipher, especially for non-technical employees. Attempting to overcome this, we sought to build an abstract layer on top of the data that allowed users to input conversational language descriptions and display corresponding parts across ERP systems. Easier said than done. Not only was the technology for natural language processing (NLP) still maturing, but there were troves of data that were not yet digitized (e.g., paper drawings), and image recognition and processing at the time were nascent, at best.
Ultimately, while we did capture some value, the bolder aspiration to use several types of automation to optimize procurement came up short. The technology was not mature enough to make our vision a reality. Even still, it was a good idea, and it remains a good idea. I am not the only executive to have had this experience. Many other examples are found across industries and enterprises, and therein is the opportunity. Today, technology capabilities offer far more sophisticated functions than those in years past, and this offers another avenue for (re)discovering where GenAI and agentic systems can generate value.
Tapping Untapped Sources Of Ideas
Many leaders who pushed the boundaries of technology capabilities years ago are in senior executive positions today. Their institutional and historical knowledge is fodder for thinking about AI applications that do more than automate a discrete part of an existing process. Indeed, the most transformational opportunities with GenAI and agentic systems may have already been identified.
With this perspective, the next steps take place in the C-suite. Executives at leading companies are already collaborating to pursue AI value. In the deliberations over strategic goals, capital deployment and technology transformation, leaders should think back on prior technology demonstrations and pilots—perhaps even at another enterprise—and reconsider whether those good ideas still hold merit.
As well as consulting their own experiences, executives can turn to their business units to solicit ideas from career employees and review documentation from previous projects. Board members may also be a source of opportunities and ideas, informed by their knowledge of the business and their own leadership experiences.
The core insight for leaders is to keep an eye on technology advances. We are in a transformational moment with the growing maturity of GenAI and agentic systems. Valuable ideas come from everywhere, including the past.
Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?
