Most large companies need significant infrastructure upgrades before they can effectively deploy artificial intelligence agents across their organizations, according to new research from Tray.ai that highlights the growing pains in enterprise AI adoption.
A survey of over 1,000 enterprise technology leaders and practitioners found that 86% of companies require tech stack improvements to properly implement AI agents. The study, commissioned by integration platform provider Tray.ai, revealed that 42% of enterprises need to connect to at least eight different data sources to successfully deploy these AI systems.
Despite these challenges, companies are betting big on AI agents, with 42% planning to build more than 100 prototypes and 68% allocating $500,000 or more annually to these initiatives. However, nearly half of respondents reported their existing integration platforms are only “somewhat ready” for AI’s demanding data requirements.
Security emerged as the primary concern across both leadership (53%) and practitioners (62%), followed by data governance and integration complexity. The findings suggest a potential disconnect between ambitious AI deployment goals and organizational readiness, with 64% of enterprises wanting three-week deployment timeframes despite current implementation delays.
“We’re seeing a concerning pattern in enterprise AI implementation that is reminiscent of the early days of cloud adoption,” said Rich Waldron, co-founder and CEO of Tray.ai. “Organizations clearly understand that seamless integration is essential, yet many are opting for patchwork approaches that will prove costly down the line.”
Companies are currently pursuing various implementation strategies, including hybrid approaches combining build and buy solutions (41%), single-purpose SaaS app agents (28%), and custom development (24%). Top use cases focus on IT service desk automation (61%), data processing and analytics (40%), and code development and testing (36%).
The survey, which included responses from C-level executives to practitioners at organizations with 1,000 or more employees, found that enterprises remain optimistic about AI agent adoption. More than two-thirds of organizations expect AI agents to power more than a quarter of their core processes by 2025, with key goals including improved efficiency, enhanced productivity, and increased customer satisfaction.
