How to Bridge the Operational AI Gap
▼ Summary
– A survey of 500 senior IT leaders found that organizations with a strong integration foundation are more likely to have advanced, enterprise-wide AI implementations.
– The research shows tangible AI progress, with 76% of surveyed companies having at least one department using a fully operational AI workflow in production.
– AI implementations are most successful when applied to well-defined and automated processes, with 43% of organizations finding success this way.
– Most organizations lack dedicated AI teams, as only 34% have a specific team for maintaining AI workflows, while responsibility is often dispersed.
– Companies using enterprise-wide integration platforms are five times more likely to use diverse data sources and have more multi-departmental, autonomous AI implementations.
A recent survey of 500 senior IT leaders at mid-sized and large U.S. companies reveals a clear path forward for deploying artificial intelligence. The findings indicate that while many organizations are now moving AI projects into active use, the most significant advancements are tied directly to a company’s underlying technological infrastructure. A strong integration foundation is directly linked to more sophisticated and widespread AI implementations, enabling initiatives that span entire enterprises. As AI tools become more common and complex, a dedicated integration platform helps prevent redundant efforts, breaks down data silos, and provides essential oversight for increasingly autonomous systems.
The research shows tangible progress is being made. After years of reports highlighting stalled AI initiatives, the data presents a more optimistic picture. Three out of four organizations (76%) now have at least one department running a fully operational AI workflow in production. This marks a significant shift from theoretical exploration to practical application within many businesses.
Success appears to follow a predictable pattern. AI implementations deliver the best results when applied to processes that are already clearly defined and automated. Close to half of surveyed leaders (43%) reported success using AI in this manner. A further quarter found success by applying AI to newly created processes, while about one-third (32%) are experimenting with AI across a variety of different workflows.
A notable structural gap persists, however. Two-thirds of organizations lack a dedicated team focused solely on maintaining AI systems. Only 34% have established such a specialized group. Maintenance responsibilities often fall to existing departments: central IT handles it for 21% of companies, and departmental operations teams manage it for 25%. In 19% of organizations, the duty is distributed across various groups without a clear owner.
The most compelling finding centers on the power of integration. Companies that have invested in an enterprise-wide integration platform are achieving far more robust AI outcomes. These organizations are five times more likely to incorporate a diverse range of data sources into their AI workflows. Specifically, 59% of them use five or more data sources, compared to just 11% of companies using integration for only specific workflows and 0% of those without any integration platform.
The benefits extend beyond data. Firms with integration platforms also demonstrate more multi-departmental AI deployment, grant greater autonomy to their AI workflows, and express higher confidence in increasing that autonomy in the future. This suggests that a cohesive technological backbone is not just a support tool but a critical enabler for scaling AI responsibly and effectively across an organization.
(Source: Technology Review)
