The Payoff: Turning 2024 Data & AI Investments ROI Positive in 2025

e notes that "only 5% or 6% of large organizations have any production deployment of generative AI," with the majority still in experimental phases. This experimentation, while valuable for learning, has yet to translate into significant economic returns. The panelists agree that AI has not yet reached its full potential, with Davenport suggesting that the industry is in the "early stages" of realizing ROI from AI investments. This sentiment is echoed by Daniel Houm, who emphasizes the need for organizations to move beyond experimentation to achieve meaningful outcomes.

Industry-Specific Applications

The panelists explore the application of generative AI across various industries, highlighting its transformative potential in sectors like content creation and customer operations. Daniel Houm shares insights from WPP, where AI is used to create content and test audience reactions rapidly, thus enhancing marketing efficiency. Houm describes generative AI as unlocking two critical parts of the marketing supply chain: content creation and testing. However, he also notes the challenge of measuring the quality and value of AI-generated content, a sentiment shared by other panelists who stress the importance of finding the right algorithms to optimize different supply chain aspects. Michelle Bonnet adds that in financial services, AI helps improve internal processes, enhancing efficiency and reducing time to resolution for customer queries.

Frameworks for Evaluating AI Use Cases

A significant portion of the discussion is dedicated to methods for evaluating AI use cases, highlighting the importance of prioritization. Michelle Bonnet suggests a structured approach, recommending that organizations assess potential projects based on impact, risk, and data availability. She advises focusing on high-impact projects, considering risk factors, and ensuring data readiness. This approach, she argues, helps organizations make informed decisions about where to allocate resources and how to maximize the ROI of AI initiatives. The panelists agree that ethical considerations should also be integrated into this method, with Daniel Houm emphasizing the need for explainable AI that is auditable and transparent.

Future of AI and Industry Impact

Looking ahead, the panelists discuss the future trajectory of AI technologies and their potential impact on various industries. Daniel Houm envisions a rapid evolution in AI capabilities, predicting that within 18 months, AI could achieve a level of reasoning akin to a PhD, eventually leading to Artificial General Intelligence (AGI). He cautions, however, that the speed of industry adoption might lag behind technological advancements. Michelle Bonnet highlights the potential for smaller, more efficient AI models and innovations in active inference to drive ROI. The panelists agree that staying informed and strategically using emerging technologies will be vital for organizations seeking to maintain a competitive edge in an AI-driven environment.