Businesses that fail to keep pace with artificial intelligence (AI) adoption risk losing an average of 8.6% of their monthly revenue, according to a study from data platform developer Couchbase. For large enterprises, this could translate to almost $87 million in annual losses.
The Couchbase survey covered 800 senior IT decision-makers from companies with at least 1,000 employees across industries including finance, healthcare, and gaming. It found that many enterprises are struggling with AI control and readiness. About 21% admitted to having “zero” or “insufficient” control over AI use, exposing them to risks from either excessive or restricted access to tools. About 64% said they are moving too slowly due to “decision paralysis.”
“The evolution from GenAI to agentic AI is creating vast opportunities for enterprises that can harness these technologies effectively,” said Julie Irish, chief information officer at Couchbase. “Creating and operating innovative AI applications at scale is essential for successful enterprises.”
Irish emphasized that a strong data strategy, focused on quality, scalability, and accessibility, is necessary for businesses to fully benefit from AI.
The survey also revealed that 78% of respondents believe early adopters of AI will become industry leaders, while 73% said AI is already transforming their technology environment. With this urgency, AI spending on technologies such as generative AI and agentic AI is expected to grow by 51% between 2025 and 2026, outpacing the 35% growth forecast for overall digital modernization. AI will account for more than half of modernization investments during that period.
However, adoption challenges remain widespread. Almost all respondents, or 99%, said they encountered issues that disrupted or halted AI projects. These included difficulties in managing or accessing required data, fears of failure, and budget constraints. Such challenges consumed 17% of AI investments and delayed strategic objectives by an average of six months.
A lack of understanding of data requirements emerged as a major obstacle. About 70% of enterprises admitted they had an “incomplete” grasp of the quality and real-time accessibility of the data needed to support AI. This lack of knowledge contributed to 62% not fully understanding the risks tied to AI, such as security and data management concerns.
On the other hand, enterprises with stronger data awareness reported greater confidence in AI adoption. They were also 33% more likely to be prepared for agentic AI compared to those with weaker understanding.