Market research firm Gartner has projected that by the close of 2025, approximately 30% of generative AI initiatives will be shelved by companies following their proof-of-concept phases. The tech research firm attributes this outcome to issues surrounding data quality, insufficient risk management, mounting costs, and ambiguous business value.
Economic and Operational Challenges
Applications of generative AI, including coding assistants and document search tools with Retrieval-Augmented Generation (RAG), are facing substantial economic and operational hurdles. Initial costs for coding assistants can range from $100,000 to $200,000, with ongoing annual expenses per user between $280 and $550. More complex AI models, such as those in healthcare or finance, involve initial investments from $8 million to $20 million and recurring yearly costs between $11,000 and $21,000.
Rita Sallam, Distinguished VP Analyst at Gartner, pointed out the financial constraints linked with developing and implementing generative AI. Speaking at the Gartner Data & Analytics Summit in Sydney, Sallam mentioned that “executives are eager to see quick returns,” although it is challenging to demonstrate tangible value early on. She remarked that costs, risks, and strategic impacts vary significantly depending on the specific applications and implementation strategies.
Proof-of-Concept Barriers
The proof-of-concept stage serves as a crucial point for assessing AI project feasibility. Nevertheless, numerous projects stall at this phase due to earlier mentioned challenges. Gartner’s research suggests that as language models advance, companies may reconsider and find new potential uses for generative AI technologies.
A recent Gartner survey of 822 business leaders, held between September and November 2023, indicated that early users of generative AI saw a 15.8% increase in revenue, a 15.2% reduction in costs, and a 22.6% boost in productivity. Despite these gains, generative AI demands a higher tolerance for future-focused financial investment criteria, a point where many CFOs have historically been hesitant.
Future Outlook and Strategic Guidance
Gartner’s analysis points to the need for organizations to thoroughly scrutinize the costs and prospective benefits of generative AI initiatives. Though the technology promises substantial potential, addressing financial and operational hurdles is essential for realizing its value. As language models become more advanced, there could be renewed interest in exploring their capabilities across multiple sectors.
Last Updated on November 7, 2024 3:28 pm CET
