Why 61% of Manufacturers Still Haven't Fully Deployed AI and How to Close the Gap
Key Highlights
- Rigid org structures are stalling AI adoption. Cross-functional workflows are now a competitive necessity, not a nice-to-have when it comes to AI.
- Supply chain intelligence is moving in-house, giving manufacturers the ability to model scenarios and respond to disruption in real time.
- AI is turning sustainability from a reporting exercise into a continuous feedback loop embedded directly into daily operations.
According to NIST, only 39% of manufacturers have fully deployed AI in their production operations. And though manufacturers are ramping up AI adoption this year, it seems that widespread use has not yet translated into enterprise-wide transformation.
There’s a clear gap between ambition and execution.
Success of AI deployments depends on manufacturers knowing where to start, how to scale and how to achieve tangible results. AI is set to revolutionize the shop floor but only for those organizations willing to make changes across four critical areas.
AI workflows demand a rethink of traditional structures
Almost all manufacturing businesses are built around rigid hierarchies, siloed departments and sequential workflows. Today, those same structures are holding many companies back. AI can link planning, production, supply chain, service and workforce activity in real time, but in organizations still designed for linear processes, its potential stalls between departments. Intelligence gets trapped in functions and progress moves at the pace of approvals, not technology.
Instead of relying on periodic, external analyses from third-parties or consultants, manufacturers will start using AI-enabled supply chain intelligence tools internally on a regular basis to explore scenarios, test assumptions and respond more quickly to change.
To move forward, manufacturers need to rethink their design, not to cut roles or create new org charts, but to break the barriers that keep AI from delivering its full value. Real returns will come to the organizations that move beyond outdated hierarchies and create systems where work, decisions and outcomes flow freely across functions to support new levels of speed, clarity, and performance.
Automation will lighten the workload
Even after significant investment in digital transformation efforts, many manufacturers are questioning why output hasn’t caught up. The biggest constraint is capacity as labor shortages reach a breaking point. The Manufacturing Institute projects that 2.1 million roles could go vacant by 2030 at a cost of up to $1 trillion in lost output each year.
Skilled technicians are retiring faster than replacements enter the workforce, and open roles remain unfilled for months. In factories already running lean, every vacancy compounds downtime and lost throughput.
It will require manufacturers to rethink daily workflows, update safety protocols and create teams that can move fast and confidently alongside intelligent machines, which requires an understanding of which tasks will be handled by people and which will be handled by machines.
This is why the next step in industrial productivity won’t come from machines replacing people, it will come from them working as a team, alongside AI-enabled systems. It will require manufacturers to rethink daily workflows, update safety protocols and create teams that can move fast and confidently alongside intelligent machines, which requires an understanding of which tasks will be handled by people and which will be handled by machines.
Prioritize in-house intelligence for more resilient supply chains
Supply chain data often remains scattered across multiple systems and formats. This challenge isn’t going away any time soon, but manufacturers need to begin changing how they leverage this data. AI can now extract, organize and make sense of this disparate data, even when it’s siloed or inconsistent.
For instance, AI-enabled supply chain modeling and simulation tools can help manufacturers use this data, even where gaps exist, to build and test scenarios across the entire supply chain.
In 2026, I expect supply chain intelligence to become an in-house capability. Instead of relying on periodic, external analyses from third-parties or consultants, manufacturers will start using AI-enabled supply chain intelligence tools internally on a regular basis to explore scenarios, test assumptions and respond more quickly to change.
Over time, this approach will embed optimization, resilience and value creation directly into day-to-day supply chain management, turning intelligence from a one-off exercise into a continuous operational advantage.
Treat environmental performance like a KPI
Manufacturers can no longer treat environmental performance as an afterthought. Instead, it must be measured with the same discipline applied to cost and quality.
Sustainability is poised to become AI-enabled and embedded into how factories, supply chains, workforces and assets are managed day to day. It will be integrated directly into planning, execution and optimization cycles.
AI systems can bring fragmented data together, track resource use at the source and provide real-time insight into energy consumption, emissions and waste. Tasks that once relied on lengthy reporting cycles or audits will evolve into a continuous feedback system—one that learns, detects anomalies and guides adjustments before targets are missed.
About the Author
Maggie Slowik
Maggie Slowik is global industry director for manufacturing at IFS.
