By the end of 2026, manufacturers are projected to spend $16.7 billion on artificial intelligence (AI). Many manufacturers have approached AI with the mindset “if we build it, they will come,” which, in this context, becomes “if we buy it, it will get used.”
However, the 2026 Asset Lifecycle Report from Asset Management Software, Siemens shows that while adoption is high, organizations are struggling to integrate AI into the core of facilities work, such as tracking assets. Sixty-two percent of manufacturers are already using AI in asset management, 77% believe it will shape their future maintenance strategy, and 75% consider it necessary to remain competitive. Yet only 51% feel prepared to shift toward predictive maintenance.
IBM suggests that AI in manufacturing currently sits squarely in document summarization, product searches, and call processing—areas that support the manufacturing process rather than sit inside it. If AI remains on the sidelines rather than becoming central to how facilities leaders track and monitor assets, manufacturers remain exposed to the same problems AI is intended to solve.
The slow shift toward predictive maintenance is not due to a lack of technology. The industry already has the necessary tools and has invested significantly to reach this point. The real challenge is preparing the people. Manufacturing leaders aiming to move their organizations to AI-informed predictive maintenance models must focus on their workforce. Facilities leaders must treat workforce readiness with the same importance as technology readiness, building trust and fluency in AI at the same pace as its deployment.
The Predictive Imperative
Historically, manufacturing maintenance has followed two main models. The first is reactive, where repairs are made only after equipment fails. The second is preventive, scheduling maintenance based on time or usage thresholds. While preventive maintenance is an improvement over a reactive approach, it can result in unnecessary work. Facilities and plants often replace or repair assets too late, potentially impacting performance and quality, or sooner than necessary. Both resulting in wasted resources and higher costs.
Predictive maintenance introduces an approach that uses real-time process and sensor data, together with analytics, to identify signs of failure before they become critical. When the cost of downtime can reach $125,000 for every hour production is stopped, the stakes for preventing failure are clear.
Facilities leaders who invest in predictive capabilities are setting up the data and operational infrastructure needed for the next generation of AI. Advanced systems like causal AI, which can diagnose the reasons behind failures and act, rely on comprehensive, long-term asset health data.
Both predictive and prescriptive maintenance draw on the same foundation: the asset health data, failure pattern histories, and real-time condition baselines that only a mature predictive system builds over time.
The People Problems
When it comes to preparing the manufacturing workforce for future advances in AI, manufacturing leaders face two people-related problems. First, the impending silver tsunami threatens both institutional knowledge and labor supply. Second, manufacturers have heavily invested in technology while largely neglecting investment in their people. Prioritizing technology over workforce preparation jeopardizes both the success of AI adoption and the returns manufacturers expect.
The silver tsunami refers to the upcoming wave of baby boomer retirements. According to the 2025 Bureau of Labor Statistics Current Population Survey, 2.94 million manufacturing workers are age 55 or older. When a large cohort of employees retire en masse, they don’t just withdraw from their 401(k)s; they take with them decades of floor-level experience that is foundational to how manufacturing facilities operate and maintain their assets. Older generations of manufacturing employees learned by doing, and workers could often tell how a machine was performing simply by sounds, vibrations, temperatures or environment, or anything that differed from learned norms.
AI-powered predictive approaches to facilities maintenance can help fill the knowledge gaps left in the wake of the silver tsunami. Modern asset management systems often contain decades of historical maintenance data, including failure patterns, repair histories, and asset performance over time. With predictive maintenance, manufacturing facilities can codify the institutional knowledge that would otherwise be lost to retirement.
However, empowering the younger generation of manufacturing workers to collaborate with technology requires significant investment not only in technology but also in people and in how they learn. Unlike older generations, digitally native younger workers are comfortable learning independently through online videos or digital documents.
Yet, the industry has not adequately invested in empowering that learning. Only 14% of frontline employees have received relevant training, even though 86% want to understand how AI will shape their roles. When manufacturing leaders neglect workforce preparation for AI, they undermine the technology investments they’ve already made.
When teams trust AI, feel engaged, and understand its insights, they act on alerts, continuously improve workflows, and prevent failures before they escalate. When they don’t, the technology suffers, hindering digital transformation. AI initiatives in manufacturing often fail not because the technology is not ready, but because people are not adequately prepared, engaged, or incentivized.
What This Moment Asks of Manufacturing Leaders
The manufacturing industry has funded technology, invested in AI, and set the expectation that the investment will eventually pay off. What it failed to do is prepare the people expected to make AI work. To prepare both the people and the technology, the manufacturing industry must put people first. To do that, two things must run in parallel:
- Start the journey to predictive maintenance now. The predictive maintenance insights generated—baselines, failure patterns, real-time conditions—are the same infrastructure that prescriptive and AI will require.
- Invest in ongoing workforce training. Frontline workers need sustained education on what technology is doing for them and why, and direct communication from leadership on what it isn’t doing, namely, replacing them.
To ensure every cent of money invested in the industry is an investment, manufacturing leaders must implement predictive maintenance now and ensure AI acts as a human-centric tool, designed for the good of informed workers, enhancing jobs, improving performance, and delivering outstanding products.
Kevin Clark is vice president of industrial strategy at Asset Management Software, Siemens.