The frenzy around ChatGPT has stabilized for the millions of experimental end-users. That will, however, change with the launch of GPT-4 next week. Interacting with GPT is a very interesting experience. You will find people aligned to various camps regarding opinions on the tool. Some believe the tool is dumb since it often fails to answer fundamental questions. I am in the camp that believes that generative AI technology is very powerful if you are not obsessed with the technology not being able to perform some foundational math. I think about it from the perspective of the impact generative AI can make on the way businesses run. And as you may have seen from numerous use cases being shared, the opportunities to leverage this technology is only limited by your creativity and imagination. However, the game has just begun for the technology giants who will now be competing in this space. Microsoft is ahead of the game, which gives it the advantage of integrating generative AI technology in its various applications. However, the true value of doing this must be generated by helping users of these technologies leverage this new functionality to erase conventional constraints. The pace at which multiple generative AI solutions have been launched in the last few months validates the point I consistently make that every technology will eventually become a commodity. The generative AI space will be highly competitive, but the competition will come from something other than product differentiation. Whether it is ChatGPT, BARD, Einstein GPT, or the generative AI model SAP plans to introduce, the value of these tools will lie in the solutions they can help create. The key is to think beyond automation. Most of the use cases you see are around automation; the real value, however, lies in leveraging the deep learning algorithm embedded in the tool for more advanced applications. The real opportunities lie in leveraging generative AI-based applications to transform how businesses run. SAP is also going to come up with its generative AI tool soon. Imagine the ways it can augment SAP's platform and product offerings. The most recent product, SAP Datasphere, is an excellent example of an SAP tool that can become 2X more powerful by strategically embedding generative AI capabilities in an already robust solution. The benefit then propagates to thousands of SAP customers across the globe. Uses of generative AI are plenty, and the technology can be leveraged for many business operations. Analytics, including AI-enabled algorithms are being rapidly incorporated in businesses worldwide. This trend resonates with SAPinsider’s recently published research, "Building Resilient and Agile Supply Chains Leveraging Data, Analytics and Automation," which focuses on how tools, data analytics, and automation can help transform supply chain operations. Generative AI can help across all these pillars. We will cover all these three pillars in a series of articles. This article focuses on the data pillar and provides a high-level view of how generative AI can help organizations transform their supply chain data and data management. Our most recent research, "Building Resilient and Agile Supply Chains Leveraging Data, Analytics and Automation," revealed that building a data foundation is a critical first step for SAPinsiders in their supply chain analytics journey. Complex data landscapes are created in enterprises as businesses, and their supply chains generate a large, distributed volume of essential data. It takes months for data technology professionals to understand such a complex landscape. However, technologies like generative AI can help deconstruct that complexity. Generative AI can help streamline and automate DevOps, AIOps, or MLOps processes; AI-powered bots can covertly automate operations like spinning vast application programming interface layers. Generative AI can potentially help in the following areas: Data discovery: Discovering master data and identifying data domains becomes challenging as the volume of data by supply chains explodes. This gets exacerbated by the rapidly expanding portfolio of applications within large organizations. A study by Okta found that, on average, large companies have 175 applications, and smaller companies have 73. Generative AI can help automate data discovery and label massive supply chain data. AI-enabled tools that can help you automate these processes are already available in the market. But the great opportunity with open source tools is that solution you build addresses the unique nuances of your supply chain data. An example is spend management. Building the spend cube is the core of an effective spend management solution. Generative AI can help develop the spend cube and keep the data therein harmonized and accurate. Data lineage: As organizations develop a single source of truth through data integration or centralization, they must also map master data movement across applications and sources within the enterprise. Generative AI-based tools can help perform technical metadata scanning and use machine learning-based discovery of relationships to map lineage, which is critical for the master data flow view. As you can imagine, data lineage becomes much more critical in the context of supply chain data. A typical supply chain portfolio of a large company is akin to a digital Frankenstein due to a plethora of legacy point systems and platforms. As organizations attempt building single-source-of-truth amidst this complexity, generative AI tools can ensure that the single-source-of-truth has the correct data from the right sources and is being leveraged by the right application. Data modeling: A single source of truth master data management hub used by applications, supply chain solutions, and analytics data stores can help organizations leverage supply chain MDM for operational and analytics use. A data model with attributes and hierarchies to create that hub needs to be developed. These attributes and hierarchies must be consistent and uniform across all the data sources. Generative AI solutions can automate schema mapping, thereby helping to recommend core attributes and hierarchies that can be used for data models. Again, many AI based solutions already exist but you can leverage generative AI to build customized solutions. Data creation: Data science initiatives in supply chains may need synthetic data for many purposes. Synthetic data generation can be helpful when limited data files are available due to the time constraint of pilot projects or when the algorithm uses a particular data format. Generative AI can create synthetic data for analytics and modeling purposes, fine-tune algorithms automatically, and generate synthetic data, which is as good, if not better, than real-world data.