76% of companies have started production-level AI, but 90% of those that are successful rely on integration platforms to operationalize their AI initiatives
Celigo, the intelligent automation platform built for AI, announced the publication of a new report in partnership with MIT Technology Review Insights entitled, Bridging the Operational AI Gap. Based on a December 2025 survey of 500 senior IT leaders at U.S. companies, the report reveals that while the majority of enterprises have moved into production-level AI, a massive operational gap exists for those attempting to scale without a unified integration strategy.
Also Read: CIO Influence Interview With Jake Mosey, Chief Product Officer at Recast
The Integration Mandate for AI Success
The findings identify a near-universal commonality among organizations that have successfully moved AI beyond the pilot phase. Instead of treating AI as a standalone tool, it is an integrated part of the business stack. According to the data:
- 90% of organizations with at least one AI workflow fully in production already utilize an integration platform; and
- 39% with enterprise-wide integration have deployed AI across multiple departments. In stark contrast, 1% or less of organizations without an integration platform have been able to scale AI beyond a single department
“Enterprises are quickly realizing that AI strategy is really architecture strategy,” said Ronen Vengosh, Chief Strategy Officer at Celigo. “The problem is that most business processes span multiple systems, but if you’re using AI tools within your CRM or ERP, for example, the information comes from within those applications. If AI cannot see across systems, it cannot reason across them.”
The Data Complexity Bottleneck
As AI systems become increasingly autonomous, the report highlights integration platforms as a foundational layer that enables consistent data access, cross-system orchestration and governance. Without this layer, organizations remain trapped in silos:
- 59% of enterprise-wide integration platform users draw from five or more data sources for their AI workflows; and
- Remarkably, 0% of organizations without an integration platform were able to reach the same level of data complexity, limiting their AI to basic, single-source tasks
The Path to an Autonomous Enterprise
Beyond initial deployment, the report indicates that leaders are moving into the next phase of maturity with mostly autonomous workflows. In these high-growth environments, the integration platform has transitioned into a vital governance layer that ensures AI remains reliable and resilient to changing business needs:
- 34% of organizations with enterprise-wide integration have already achieved mostly autonomous AI workflows; and
- 95% of executives say their AI workflows already have some level of autonomy, with most expecting autonomy to increase over the next 12 to 18 months
“We are seeing a clear shift from AI experimentation to a focus on the infrastructure required for scale,” said Jan Arendtsz, Founder and CEO of Celigo. “This report proves that to move toward agentic workflows that drive actual value, companies must bridge the gap between their disparate applications and their AI models.”
Catch more CIO Insights: Why CIOs are becoming chief risk orchestrators?
[To share your insights with us, please write to psen@itechseries.com ]
