Published on December 15, 2023, 3:09 pm

Generative AI: The Slow Adoption and Tension Faced by Enterprise Buyers

Generative AI, a form of Artificial Intelligence (AI) that uses large language models to generate human-like text, has been gaining significant attention from vendors. They claim that this technology can revolutionize the way companies operate and help them be more efficient. However, despite the hype, enterprise buyers have been cautious in their approach.

Large companies tend to be prudent when adopting newer technologies. While generative AI holds great potential, it also comes with its own set of costs. For instance, implementing generative AI features in Software-as-a-Service (SaaS) products may result in higher expenses. Similarly, building internal software using large language model APIs can also be costly.

For this reason, companies want to ensure they get a return on their investment before fully committing to generative AI projects. According to a July survey of large company CIOs conducted by Morgan Stanley, while 56% of respondents acknowledged the impact of generative AI on their investment priorities, only 4% had launched significant projects. Most companies were still evaluating or experimenting with proof of concepts.

Nevertheless, despite the cautious approach taken by CIOs and CTOs, there is pressure to deliver exceptional experiences similar to those provided by popular online tools like ChatGPT. Internal customers within organizations have experienced the capabilities of these tools and expect similar results from their own environments.

To handle this tension between delivering what customers desire and ensuring careful implementation, companies must establish proper structure and organization. Jim Rowan from Deloitte suggests focusing on infrastructure beyond just technology. This includes setting up processes, governance, identifying key personnel involved in implementation, and discussing use cases to address specific business problems.

Several CIOs we spoke with have adopted a similar strategy for implementing generative AI within their organizations. Monica Caldas at Liberty Mutual started with a proof of concept involving a few thousand employees and plans to expand it across the company’s 45,000-strong workforce. Mike Haney at Battelle has been exploring various use cases for generative AI and has seen positive results in terms of efficiency improvement. Kathy Kay from Principal Financial Group initiated a study group where engineers and business professionals curated around 25 use cases, with three already moving into production. Sharon Mandell at Juniper Networks is participating in an initial pilot with Microsoft to explore the benefits of its generative AI tool, Copilot for Office 365.

Despite the interest and potential power of generative AI, companies understand the need to approach its implementation carefully. As executives learn more about this transformative technology through experimentation, they strive to strike a balance between meeting customer expectations and managing risks associated with early adoption.

In conclusion, while vendors may hype up generative AI, enterprise buyers are rightly cautious. They recognize that implementing such a technology requires careful consideration of cost-effectiveness, infrastructure setup, and understanding use cases that will truly drive organizational efficiencies. The slow adoption of generative AI by enterprise buyers reflects their pragmatic approach to new technologies.


Comments are closed.