Volume 29 | Issue 2
Click here to read the complete illustrated article or continue below to read the text article.
By Shane Kehoe, Global Head of Practice, Manufacturing & Distribution, World Wide Technology
Manufacturers are investing heavily in AI, but most remain stuck in pilot mode. The challenge is not a lack of ambition, funding, or AI capability — it is operational readiness.
While manufacturers are deploying AI across predictive maintenance, computer vision and supply chain forecasting, most organizations have not yet scaled these initiatives enterprise- wide. McKinsey research shows that nearly two-thirds of organizations have not begun scaling AI across the enterprise, with only about one-third report reaching a scaling stage.
This scaling gap is also evident in manufacturing specifically. Deloitte’s 2025 Smart Manufacturing and Operations Survey finds that only 29% of manufacturers have deployed AI and machine learning at scale across facilities, despite widespread investment in smart manufacturing initiatives. At the same time, McKinsey highlights that the gap between digital and AI leaders and laggards has widened by approximately 60% over the past three years, reflecting a compounding advantage among more mature organizations.
Based on World Wide Technology’s work with more than 100 global manufacturers, the organizations seeing the greatest success are not necessarily those with the most advanced AI models, but those investing first in the data, integration and operational foundations required to scale AI reliably across the enterprise.
Across WWT’s manufacturing engagements, a consistent pattern is emerging. Manufacturers are not struggling to identify AI use cases or secure funding. They are struggling to create the operational conditions and data architecture required to scale use cases beyond pilots. The challenges are clear: data and integration foundations are weak, ROI and ownership are unclear and plants are not ready to adopt pilots into real workflows.
Most manufacturers run 3-5 AI pilots simultaneously but fewer than 30% ever scale to production. This gap between proof of concept and production value is where ROI is not realized. The lack of production scale and adoption is comparable to IIoT pilots within Manufacturing environments previously.
The root causes are:
According to WWT’s sector analysis, these challenges must be addressed before AI can scale effectively. Technology alone cannot overcome foundational integration gaps.
“The real AI leaders in manufacturing aren’t running the most pilots – they’re building the strongest operational backbone with cross business unit buy-in. Without true IT/OT convergence, unified data layer and shared commitment across functions, even the most sophisticated AI models stay stuck as isolated experiments instead of scalable, production systems.”
– Shane Kehoe, Global Head of Practice, Manufacturing & Distribution, World Wide Technology
Across many manufacturers, AI adoption is increasing rapidly. However, most deployments remain confined to individual sites or functions. Manufacturers are successfully launching isolated pilots such as predictive maintenance in one plant or quality inspection in another. The challenge comes when attempting to replicate those results across multiple environments.
The limitation is not the AI itself. It is the supporting infrastructure, including:
Successful scaling requires treating infrastructure and organizational alignment as core components of AI strategy, not secondary considerations.
WWT consistently identifies data fragmentation as the primary barrier to scalable AI in Manufacturing. Many organizations operate with:
While manufacturers often invest in data infrastructure before launching a pilot, they typically address only one data stream for a single use case. True scaling, however, demands data architecture that considers integrating dozens of systems, many of them decades old, that were never designed to work together. This fundamental challenge becomes apparent when attempting to deploy solutions across complex environments. A predictive maintenance model trained on three months of carefully curated sensor data from a single machine, for instance, cannot reliably generalize to a plant housing 400 machines of varying ages, with inconsistent tagging conventions and data gaps resulting from unplanned downtime. The result is that while pilots may appear successful, the underlying data problem isn’t truly solved. It’s simply deferred to the scaling phase.
Without unified and reliable data pipelines, AI systems cannot deliver consistent outcomes across the enterprise.
One of the most consistent findings from WWT’s manufacturing work is that IT/OT convergence is essential for AI scalability.
IT wants to move AI workloads to cloud infrastructure, while OT prioritizes uptime and is deeply resistant to anything that introduces latency or connectivity dependencies on the production line. Security policies conflict. Change-management processes conflict. Even when a pilot succeeds technically, the path to production deployment often stalls at the governance level. Critical questions emerge: who owns the system? Who bears responsibility when the AI makes a wrong call on a production line? These questions are rarely answered before a pilot begins, yet they routinely block scaling efforts indefinitely.
Manufacturers cannot effectively scale AI without connecting operational systems with enterprise IT platforms; modernized infrastructure, security and data contextualization provides the foundation for industrial AI at scale.
This is no longer just an IT initiative. It is a business transformation requirement spanning operations, supply chain, quality and leadership.
WWT’s experience shows that successful organizations do not attempt to scale AI immediately. Instead, they follow structured progression.
Manufacturers that scale AI successfully treat the pilot not as the goal but as a diagnostic tool. They invest in data understanding and architecture first, defining commercial success metrics and planning for change management as a core workstream, not an afterthought. The ones still cycling through pilots are optimizing the wrong thing:
This phased approach reduces risk and ensures that AI initiatives are grounded in operational reality. The question for Leadership is no longer “can AI work here?”; it is “are we building the conditions for it to scale?”
Across manufacturing environments, the barriers to scaling AI – disjointed data, legacy OT systems, IT/OT misalignment and inconsistent governance – are often treated as separate challenges. They are symptoms of a single underlying issue: the lack of a unified operational and data architecture.
AI does not fail at the model level in manufacturing; it fails at the integration layer between systems, processes and accountability structures. Until that layer is addressed, scaling will continue to stall regardless of model sophistication or investment levels.
Manufacturing AI adoption continues to accelerate, but scaling remains the central challenge. While experimentation is widespread, enterprise transformation depends on operational readiness.
The organizations that will lead in AI-enabled manufacturing are not those running the most pilots, but those building the strongest operational foundations.
AI success depends on connectivity, data usability and alignment across IT and operations. Without these, even the most advanced models will remain isolated experiments rather than enterprise capabilities.
World Wide Technology has worked with manufacturing leaders across the globe on AI transformation. WWT has seen what works and what doesn’t. If you’re wondering whether your organization is ready to scale AI beyond pilots, WWT can help. Learn more about WWT’s AI for Manufacturing solutions.
AI pilots often succeed in controlled environments but fail at scale due to inconsistent infrastructure and lack of standardization. WWT findings show that scaling requires integrated systems, unified data models and cross-functional alignment.
Learn more: WWT’s AI Transformation Research
Define business layer success metrics prior to assessing IT/OT integration readiness. Getting cross-business unit buy-in and ownership is critical. Conduct a data audit to understand what’s actually accessible and in what condition. Align IT and OT on governance before deployment. Pilot in realistic (not ideal) conditions. If the pilot can’t survive the real floor, don’t build a production system on its results.
Learn more: IT Modernization for Manufacturing
Automotive, Semiconductor and Lifesciences would be leading in manufacturing AI adoption, but every company is different. There is definitely “Leaders vs Laggards” emerging in AI adoption and the leading companies have a very different attitude, business focus, investment strategy and mindset. According to McKinsey, the “digital and AI maturity gap between leaders and laggards has grown 60% in 3 years and is compounding.”
WWT recommends prioritizing investments based on operational readiness and business value. Organizations should focus first on data integration and infrastructure modernization before expanding into advanced AI use cases.
About the Author:
Shane Kehoe is Global Head of Practice for Manufacturing & Distribution at World Wide Technology, where he leads strategic initiatives focused on helping industrial and manufacturing organizations accelerate digital transformation, operational efficiency and business growth. With more than 25 years of experience across manufacturing, distribution and high-tech industries, Shane brings deep expertise in industrial automation, IT/OT convergence, Industry 4.0, AI, hybrid cloud and supply chain optimization. A recognized thought leader in the future of manufacturing and automation, Shane is passionate about how advanced edge, cloud and AI technologies are reshaping industrial operations and supply chains.
Forging the Next 250 Years: Powering the Next Era of American Manufacturing
As manufacturers offer more customization than ever before, managing product complexity has become a critical challenge. Tune in with Dan Joe Barry, Vice President of Product Marketing at Configit, who explores how companies are tackling the growing number of product configurations across engineering, sales, manufacturing, and service. He explains how Configuration Lifecycle Management (CLM) helps organizations maintain a single source of truth for configuration data. The result: fewer errors, faster quoting, and the ability to deliver customized products at scale.
Get In Touch
Google news and SEO compliant, Industry Today’s state-of-the-art digital media platform offers bespoke media campaigns that target key decision makers and buyers to achieve your marketing and promotional goals.
Contribute
Showcase your brand and promote your business to our highly targeted audience. We offer detailed Google Analytics with measurable ROI to assure success. Submit your content for review by our Editorial team who will contact you to discuss the project further.
About Us
Reach Your Targeted Audience and Grow Your Business. Learn more About Industry Today.
Contact Us
© 2026 Industry today. All Rights reserved.
