Generative artificial intelligence, or Gen AI, is the branch of AI that has captured attention in recent years, bringing to the forefront a technology already known and used in the enterprise environment. Today, Gen AI is at the center of an adoption rush that, however, is not always justified, especially when measurable results are sought. Let us try to identify some uses of this technology that, already, include a measurability component.
Gen AI and artificial intelligence
To better understand the different fields of application, let us clarify the difference between general AI and generative AI.
General AI is a broad field of studies and applications, which includes systems capable of performing different tasks, from visual recognition to analytics to machine learning.
Generative AI is a subset focused on the Creation of new content: text, images, music, video or other types of files, based on the data with which it has been trained. This considerable potential must be brought to bear in a way that provides results beyond mere experimentation.
Gen AI: how to measure ROI
To measure the effectiveness of Gen AI, it is advisable to Parameter it to traditional measures, thus noting the impact on existing business processes. Using performance indicators that are too specific risks creating a self-referential effect: a system that, while working well, has no real impact on business operations.
Here are some KPIs suggested by Regesta LAB for concrete and effective evaluations:
- Direct returns: Undoubtedly this is the easiest and most immediate category of KPIs to understand and systematize. Included in this category are the time saved by operators for the same quality of output, or the actual economic savings achieved, for example, by implementing a maintenance system capable of responding naturally to technicians’ requests.
- Indirect returns: This second category includes measures that, while not having a quantifiable return in the first instance, are nevertheless part of fields canonically recognized to be value-adding. Think, for example, of improving user experience or reducing response times for requests for information or assistance.
To summarize, we can say that, with the exception of entities engaged in truly pioneering research projects, it is easy to imagine that, if it is necessary to “invent” a dedicated KPI, it is very likely that the return is not all that measurable.
Five business uses of generative AI with measurable ROIs
Having made the relevant assumptions, let’s discover some examples of uses of Gen AI that follow the simple criterion of return measurable results or provide an immediate benefit to companies.
- Generative design. Also called Generative Design, it brings AI into one of the most strategic business contexts, using machine learning algorithms to mimic the work of engineers in the design and prototyping phase. Designers input all necessary parameters, such as materials, dimensions, weight, required strength, manufacturing methodologies and cost constraints, into generative design software, deriving all possible outcomes and variances. Using this system, thousands of design options per product can be quickly generated. A tool that can offer the best performance in highly serialized supply chains.
- Document management. It is one of the most obvious capabilities of AI-based tools, allowing them to quickly extract the required information even from huge data lakes. The ability to nimbly retrieve the part number of a part that went out of production a few years earlier, or entire designs and drawings from digitized archives saves considerable time and human resources.
- Customer and user relations. We have mentioned how Gen AI is particularly suitable for creating content from scratch from a provided database. This has considerable application in all fields of first contact, particularly informational contact. A customer querying a chatbot about the hours of a specific location or the procedure for making a return will find an immediate, accurate and effective response.
- Support for operators. Equipping operators with a tool capable of quickly exploring all company documentation and providing answers that form the basis for analysis, research or problem solving will speed up time considerably and allow staff to focus on real value-added.
- Advanced maintenance management. In addition to predictive maintenance, generative AI can support unscheduled maintenance operations by querying the system about the wear status of components and necessary maintenance. The creation of a virtual assistant dedicated to maintenance ensures the reduction of time spent on lower value-added activities.
Generative AI in companies: a new tool to be evaluated
Gen AI is a tool with high potential that is beginning to prove itself fully. The examples provided make a strong case for its adoption, but careful evaluation based on selected performance indicators is always necessary. This approach is key to mediating between real business needs and market trends.
Regesta LAB, thanks to its experience alongside companies and its research and development capacity in the most innovative sectors, can help companies in evaluating and choosing the most effective adoption path.
Do you want to bring the benefits of Gen AI to your business? Tell us about your needs, our experts are ready to provide you with the answers you seek.
