A Strategic Priority Undone by Data Hoarding
Industrial leaders overwhelmingly claim that digital ecosystem strategy is central to their future. According to the new Industrial Intelligence Report by AVEVA and IMD, 74% of executives say digital ecosystems are a top strategic priority. Yet only 27% report substantial industrial data sharing with their ecosystem partners. This disconnect is stalling industrial AI integration at scale. The report frames “industrial intelligence” as the fusion of operational technology, information technology, and AI into a connected decision-making fabric. That fabric frays quickly when partners cannot access each other’s data. Integration complexity, entrenched legacy systems, and weak governance are cited as primary reasons why ambition outpaces execution. The result is a patchwork of point solutions and pilots that never graduate into fully connected, AI-ready operations—despite clear intent to collaborate more deeply across value chains.
Data Silos in Manufacturing: The Hidden AI Tax
Siloed data remains the biggest hidden tax on industrial AI integration. In many manufacturing and infrastructure environments, critical information is scattered across personal inboxes, isolated file servers, field devices, and disconnected design tools. This fragmentation undermines industrial data sharing and limits what AI systems can “see” and learn from. Egnyte’s latest capabilities highlight how pervasive the problem has become. Its Email Capture feature pulls important emails and attachments out of personal inbox silos into a governed repository, ensuring project decisions and approvals are preserved and searchable. For architecture, engineering, and construction teams, Egnyte’s integrations with tools such as Autodesk Forma and Deltek Vantagepoint connect field systems and design environments directly to a unified project record. By consolidating communications, designs, and documents into a single domain, AI tools can reason over more complete, current information—turning disconnected conversations into actionable, auditable knowledge.
Cognite and ABB Show What Connected Industrial AI Looks Like
While many organizations struggle to operationalize their digital ecosystem strategy, some industrial innovators are demonstrating what connected AI can deliver. Cognite and ABB have partnered to layer an “agentic” AI framework on top of established industrial applications such as ABB Ability SafetyInsight and AlarmInsight. Their goal: to break down data silos manufacturing operators face and orchestrate complex workflows across systems. By using Cognite’s industrial AI and data platform as a backbone, the collaboration enables agent-to-agent interactions that autonomously interpret data, apply logic, and trigger cross-system actions. Energy producer Aker BP is the first to adopt this approach as part of its push toward more efficient, scaled operations. Faster multi-system risk assessments, alarm rationalization, and improved risk mitigation are early promised outcomes, achieved by connecting previously disparate data points into a coordinated, AI-driven workflow rather than isolated dashboards.
From Strategy to Execution: Fixing Governance and Architecture
The gulf between ecosystem vision and reality is not just a technology problem; it is organizational. AVEVA and IMD’s research underscores that governance, integration, and learning capabilities are as crucial as platforms. Companies must define clear rules for industrial data sharing—what is shared, with whom, under which controls—while aligning incentives across partners to contribute data, not hoard it. On the technical side, unified data platforms such as those from Cognite and Egnyte show a path forward: standardize data models, centralize access controls, and connect operational and information systems so AI can act on consistent, context-rich information. AI connectors and model-context protocols further reduce integration friction by linking existing tools like email, messaging, and contract systems into a single industrial AI integration layer. Organizations that pair these architectures with strong governance and continuous learning loops are the ones most likely to close the execution gap and realize the full promise of connected industrial ecosystems.