A new WisdomAI survey found that 93% of enterprise data leaders are using or experimenting with AI for analytics. Yet only 7% said AI analytics has been adopted across every line of business, revealing a major gap between testing the technology and making it a standard enterprise tool.
The research surveyed 201 VP- and C-level data, analytics and AI executives at North American companies with annual revenue of at least $1 billion. The online survey was conducted in May 2026 and included respondents from healthcare, financial services, retail and software.
Overall, the survey results indicate that deploying AI analytics is both a tech challenge and, perhaps more so, an organizational challenge. Companies need to decide who maintains the business knowledge used by AI systems and how that information is governed.
Creating Context, Investing in Systems
A key barrier to greater adoption of AI data analytics is trust. A mere 19% of respondents said they are very confident in answers produced by AI. In contrast, 81% of organizations still depend primarily on dashboards and individual data requests to provide business insights. In other words, enterprises are so deeply reliant on existing dashboards and analytics processes that the shift to AI is still a work in progress.
The findings point to a thorny challenge for enterprise AI deployments: LLMs can analyze data and respond to questions, but producing a technically correct answer is only part of the task. Enterprise analytics also depends on understanding company-specific terminology, metrics, policies and operating practices, so AI systems need reliable access to this business context.
Companies are now investing in the people and systems needed to provide that context. Some 73% of respondents already have a position dedicated to developing and maintaining AI context. For organizations that lack these professionals, 33% currently have job openings for the role, while the others expect to hire within two years.
The shift to AI is reshaping existing data jobs. Half of the executives surveyed said AI has dramatically altered the roles and responsibilities of their data organizations. Over the next two years, 57% expect to reskill or upskill current data employees, while 20% plan to establish entirely new AI-focused jobs.
Companies also expect to change their underlying data operations. Some 94% of respondents plan to modify how enterprise context is stored and managed during the next 12 to 18 months.
Still, companies remain far from agreement about what AI context should include. Nearly three-quarters of surveyed organizations lack a shared definition of AI context. Among companies that have established a definition, 66% continue to maintain context across separate systems like data catalogs, semantic layers, BI software, and internal documentation.
Echoing a long-term enterprise concern, improving the underlying data remains the top priority. When asked what would improve AI accuracy, 60% of respondents selected cleaner data. Another 48% cited greater visibility into how AI produces its analysis, while 44% pointed to human feedback and 40% selected stronger governance and security.
General-purpose AI models are already part of the experimentation. Some 56% of respondents are testing LLMs such as GPT and Claude.