Above: The FG3015 sorting loader automates material loading, unloading, and sorting for a streamlined workflow.
April, 2025- From steam power to electricity and information and communications technology, change in the industrial sector is driven by innovation. The next revolution in manufacturing has arrived, stemming from the use of artificial intelligence and machine learning to analyze data and drive automated processes.
Data from the World Economic Forum shows investment in manufacturing artificial intelligence will grow 57 percent by 2026, with machines beginning to mimic aspects of human intelligence and automate complex tasks. The WEF notes that embracing human machine collaboration will introduce new possibilities for manufacturing transformation. However, about 56 percent of manufacturers are still only using AI in small-scale pilot projects, and only 28 percent have plans that have moved beyond the pilot stage.
Manufacturers around the world want to solve a list of similar problems, such as keeping up with evolving demands, navigating complex regulations like the General Data Protection Regulation and contending with global competition, says Kathleen Mitford, corporate vice president, global industry marketing at Microsoft. “Many already recognize and several are already seeing—the transformative potential of AI to address these challenges by streamlining operations, optimizing processes and enhancing products.”
STAYING AHEAD
Mitford says AI can drive innovation by accelerating product design, development and market entry; enabling intelligent factories; building resilient supply chains; and modernizing customer experience, as well as empowering engineering teams to automate design, generate real-time models and refine solutions, “giving them time back to focus more on manufacturability and regulatory compliance.” In addition, AI’s capabilities of search, summarization and speech allow business functions to boost productivity with upto-date information.
“We speak with manufacturing clients and prospects every week who are considering how to integrate AI into their businesses,” says Doug Schrock, managing principal of AI at accounting and consulting firm Crowe. “The rising interest in AI is evident in our conversations with business and IT leaders, echoing insights from our recent Technology in Metals Survey, which revealed a 70 percent increase in the adoption of AI tools by metals companies in 2024, compared to 2023.”
CHALLENGES
SCALING AI
• Shortage of specialist skills and talent
• Limitations of cloud-based computing power
• Inadequate data quality
• Cost of maintaining/improving AI models
• Inadequate governance of AI models
Source: “Taking AI to the next level in manufacturing,” MIT Technology Review Insights, Microsoft
Schrock says the most common challenges executives face when considering AI is a lack of understanding of its capabilities and uncertainty about where and how these tools can provide value. “AI projects are generally high visibility, which makes business and IT leaders hesitant to try large-scale initiatives where they lack in-house experience and few examples from industry peers to guide them.”
Mitford says manufacturers often encounter roadblocks when attempting to scale AI quickly because of data quality and governance. “By implementing technology that integrates and synchronizes OT, IT and ET data, organizations can unlock deeper insights for AI applications. This unified approach enables AI algorithms to access comprehensive, real-time data streams, enhancing predictive analytics and optimizing maintenance schedules.”
RESEARCHING POSSIBILITIES
An important first step is reviewing practical examples, which helps industrial companies identify where intelligence might seamlessly fit into their own operations. Schrock says the best use cases are ones with high amounts of unstructured data that “require human interpretation, processing and output back.”
In early 2024, meaningful AI implementations were largely within a small set of technology-forward companies, notes Schrock. “That front curve has passed, and here in 2025, we have seen a transition where most companies have now moved into a phase of do-it-yourself AI,” including individual and small-scale testing of off-the-shelf tools like Microsoft CoPilot, ChatGPT or AI specialty tools for one-off, functional use cases.
“Leaders now are those that apply the more advanced, professional-level AI tools with either in-house or external expertise,” he continues. “Last year focused on single-step generative AI, whereas this year, more companies will embrace agentic AI solutions that perform meaningful, multi-step processes to augment human teams.”
Anders Chrintz, global industry lead, manufacturing, at DXC Technology, a Fortune 500 global technology services provider, says that small manufacturing companies generally have the agility and flexibility necessary to adopt new technologies quicker. “They normally have fewer resources but they often can implement changes faster and may experiment with innovative solutions that large manufacturers might be hesitant to adopt due to their established processes.”
Smaller firms also can leverage partnerships and collaborations that allow them to access AI technology without making large capital investments, Chrintz says. And “market pressure also is something that can’t be underestimated for smaller manufacturing firms where investments in AI and smart factory are necessary to differentiate and even to survive.”
77%
Percent of firms with more than $10 billion in annual revenue that are deploying AI use cases, but just 4% of those earning between $100 and $499 million have done so.
Source: “Taking AI to the next level in manufacturing,” MIT Technology Review Insights, Microsoft
AGILITY
Using AI in manufacturing can provide numerous benefits, according to Chrintz, including increasing efficiency, predicting failures before they occur, detecting defects in production more accurately, optimizing supply chains, allowing for easier, more agile customization of products, and enabling data-driven decision-making.
Simulation is one use of AI where many manufacturers can see quick, concrete results. In late 2024, AI design specialists EvoPhase and precision metal fabricator KwikFab revealed an urban wind turbine designed by AI. Dubbed the Birmingham Blade, the turbine provides small-scale, affordable generation of clean wind energy, capturing relatively low wind speeds by managing turbulence caused by surrounding buildings. According to a press release, “Using AI was essential for breaking free of the long-standing biases that have influenced turbine designs for the past century,” and allowed the researchers to “generate, test and refine over 2,000 wind turbine designs in just a few weeks, significantly accelerating our development process and achieving what would have taken years through conventional methods.”
Simulation also is gaining ground in the architecture and construction industries, optimizing building plans, analyzing structural requirements, allowing real-time design modifications and providing cost-effective layouts using specifications. In construction and beyond, AI-enabled systems also can predict material shortages, labor needs and delays, as well as detect safety risks in real time through wearables and site sensors.
EMPOWERING PEOPLE
“AI will assist human workers by taking over repetitive tasks, allowing them to focus on more complex and value-added activities,” Chrintz predicts. However, he says that initially there may be a shortage of skilled workers who are capable of operating and maintaining AI systems.
Intelligence is not only driving the manufacturing lines themselves but also the products coming off them. At CES 2025, heavy equipment manufacturer John Deere revealed the second generation of autonomous machines, incorporating computer vision, AI and cameras, to support customers in agriculture, construction and commercial landscaping.
“Our agriculture, construction and commercial landscaping customers all have work that must get done at certain times of the day and year, yet there is not enough available and skilled labor to do the work,” commented Jahmy Hindman, chief technology officer at John Deere. “Autonomy can help address this challenge. That’s why we’re extending our technology stack to enable more machines to operate safely and autonomously in unique and complex environments.”
Deere’s 9RX tractor features 16 individual cameras arranged in pods to enable a 360- degree view of the field, as well as calculating depth more accurately at larger distances, allowing the tractor to pull more equipment and drive faster, and an electric mower for commercial landscaping leverages camera technology so staff can more easily achieve 360-degree coverage and have time to focus on other aspects of the job.
QUALITY CHECK
Historically, value chain areas like R&D have been on the cutting edge of technology use, says Microsoft’s Mitford. “We’re also seeing the automotive and electronics industries reach the AI deployment stage faster than others.”
Ford is producing cars of the future at a state-of the-art factory, particularly the all-electric Explorer. The Ford Cologne Electric Vehicle Center features cutting-edge AI and a team of 600 carefully choreographed robots that execute welding, cutting, dusting, painting and fusing tasks. The plant also uses digital twin monitors to support production line operators in delivering to quality standards. The digital twin is displayed on a giant touchscreen containing all workstations with information on tooling, material delivery and work safety.
“By monitoring and controlling every step of the manufacturing process, Ford will achieve unprecedented levels of quality for our customers,” said Rene Wolf, managing director of manufacturing at Ford-Werke GmbH.
AEROSPACE APPS
In the aviation industry, AI also is being used to produce precise components and analyze vast amounts of data to pinpoint safety risks, predict equipment failures and assist in proactive maintenance. Airbus, for example, has implemented a companywide digital transformation program that provides changes to the way products are designed, manufactured and supported, as well as reduces product development lead time and increases customer satisfaction because of late-stage customization flexibility.
GE Aerospace also is using AI to help strengthen the supply chain, digitizing 18 million past maintenance, repair and overhaul (MRO) records.
“Today, when we perform maintenance on an engine at a GE Aerospace MRO shop, key records are digitalized and AI helps verify the validity of key data fields, flagging discrepancies in real time,” according to the company.
13.2% Amount the U.S. smart manufacturing market is estimated to grow from 2024 to 2030.
BE READY AND REALISTIC
A clear AI strategy is crucial to success. Chrintz says that before kicking off a widespread AI implementation, it is important to consider the risks. “AI implementation is not a technology project. It’s a business project that needs ownership from the executive management team.”
Insufficient buy-in can lead to employees undermining the implementation of AI technologies due to fear of job loss or lack of understanding of the benefits. “A manufacturing company must develop a comprehensive strategy that includes thorough training, change management initiatives, robust cybersecurity measures and continued monitoring and evaluation of AI systems.”
He notes that an example of how the risks can affect downstream processes is if the data regarding equipment malfunctions is inaccurate, which can lead to machinery breakdowns or quality control issues, and then snowball into overstocking, stockouts, production schedule disruption, and decreased customer satisfaction.
“Well-implemented AI solutions can deliver high value, even for those organizations with outdated legacy systems and messy data,” says Schrock. “In fact, some of the best opportunities for AI arise in companies where there is a significant amount of manual work or reliance on unstructured information outside the ERP, such as PDFs, catalogs, file folders. The cost of the right AI project is often much lower than some of the historical IT projects, and the speed of implementation is typically measured in weeks, not months. This greatly minimizes risk and leads to a quicker ROI.”
When working with legacy systems, Chrintz recommends that companies conduct an assessment of the existing structure. “It is essential to find ways to work around legacy systems by adopting incremental approaches, such as using hybrid systems, cloud-based solutions or developing APIs that facilitate integration. As AI technology continues to evolve, organizations that effectively navigate these challenges gain competitive advantage.”
Few manufacturers have the budget or infrastructure to rip and replace their technologies, says Mitford. Microsoft’s modular solutions can enable manufacturers to gradually modernize their technology while extracting value. “However, what we’re hearing from customers as the bigger challenge to scale AI is the skills shortage. As digital innovation continues to skyrocket, re-skilling workers and providing cutting-edge technology is critical. It’s imperative to connect frontline employees with engineers, and more importantly, with relevant data.”
Airbus’ production line in Hamburg, Germany, uses seven-axis robotic arms that perform precise fuselage drilling.
CYBER RISKS
Security standards also are a concern. “Over the last few years, manufacturers have realized that their operational technology environments within their factories and warehouses do not have the same security standards adopted by their IT organizations,” says Mitford, noting that Microsoft has heavily invested in trustworthy, safe, secure and private AI.
There’s no way around the disruption that AI will create, however, Schrock says “if a company decides to take a technology-laggard position and not engage meaningfully with AI, they are putting themselves at a disadvantage relative to their competition,” which can create accumulating technical debt and make it difficult to catch up to more efficient competitors.
“When choosing to implement AI, organizations face common and significant risks, such as cybersecurity, ethical concerns, incorrect or misleading results, poor adoption and missed expectations,” he points out. These risks, however, can be effectively managed. “Implementing AI poorly exposes an organization to avoidable risks, while not adopting AI at all may lead to irrecoverable disadvantages,” he says.
