The manufacturing industry is on the brink of change. Technological progress, changing customer expectations, and personnel challenges demand a new approach. Manufacturers are increasingly harnessing the potential of generative artificial intelligence to optimize processes, improve efficiency, drive innovation, and enhance their competitiveness.

An international study by Capgemini, which also surveyed German companies, shows: 74 percent of manufacturing companies view it as a transformative technology that will help expand revenue and innovation in the future. Just a year earlier, in 2023, this share was 44 percent. It is therefore hardly surprising that, according to another survey by Google Cloud, almost two out of three companies in the manufacturing and automotive sectors worldwide are using generative AI in the production process. In Germany, there seems to be an awareness of the importance of AI—but regarding implementation, companies are more cautious here.

What should companies consider when it comes to implementing the technology? Here are three important trends that executives in the manufacturing sector should keep an eye on in 2025.

1. The future of manufacturing is multimodal

The multimodal capabilities of artificial intelligence have the potential to significantly change manufacturing as we know it. Multimodal AI agents can act as "guardians" in production by continuously monitoring and analyzing a variety of data sources, allowing companies to make their operations both highly efficient and effective.

Multimodal AI refers to models that are capable of processing and analyzing both structured and unstructured data across multiple types of data inputs, such as text, image and video, audio, and vibrations. This enables generative AI to offer the following:

• Descriptive insights: Insights for manufacturers on what is specifically happening in their operations

• Predictive insights: Predictions about what questions or problems might arise in the future

• Prescriptive actions: Recommendations for actions manufacturers can take to solve the predicted problems

Knowing when a machine is about to fail is invaluable for businesses. With multimodality, a system can listen to the vibrations or noises of a machine and thereby infer that the machine will soon fail. By analyzing a variety of data sources, including market trends and weather data, multimodal AI can identify patterns, correlations, and trends. This helps manufacturers to predict demand more accurately and make data-driven decisions, ultimately improving supply chain resilience.

In the coming years, more and more manufacturers will experiment with these functions and identify use cases that address numerous challenges of modern production facilities, from workplace safety to improving the customer experience.

2. Generative AI will drive customer-oriented manufacturing

In recent years, there has been a shift from brick-and-mortar retail to online shopping: customers buy everything online, from consumer goods to cars. The Postbank Digital Study 2024 shows: one in four people already predominantly shop on the internet.

This development also shapes the needs of customers, as they increasingly expect the products they seek to be as customizable as possible. This in turn leads to supply chains being forced to adapt to the needs of manufacturers—not the other way around. It is foreseeable that the traditional warehouse and sales model will evolve into a complex make-to-order sales model in the future. To drive this change forward, it is essential for manufacturers to have a reliable real-time overview of their operations. And this starts with data. Producers need to unify their IT and OT data before they can gain actionable insights through the application of analytics and AI tools, which ultimately optimize processes—from product design and production efficiency to targeted marketing initiatives and proactive customer service.

In business operations, uncertainty is the enemy of profit. Lack of transparency leads to unplanned downtime, which in turn leads to loss and creates further uncertainties in meeting demand, managing inventory, and addressing supply chain challenges. Real-time transparency in operations, made possible by OT-IT data integration, is the key to reducing this uncertainty and minimizing excess or insufficient finished products.

3. Generative AI helps bridge the skills gap in manufacturing

The shortage of skilled workers in the industry is becoming more severe as experienced employees retire and manufacturers struggle to attract and retain new talent with the necessary skills. An IW survey, which calculates the production potential using the Global Economic Model from Oxford Economics, showed: In 2024, the German economy lost production capacities worth 52.43 billion USD due to the shortage of skilled workers.

Especially in these times, it is necessary for employees on the production front to be equipped with helpful tools for real-time problem-solving. Because generative AI offers a powerful solution to the challenge of the skilled labor shortage in manufacturing by transforming the way knowledge is shared and applied. Imagine AI-powered agents capturing the expertise of experienced employees and providing new staff with personalized training programs. These agents could function as virtual mentors, offering guidance, answering questions, and providing real-time feedback.

Thanks to the automation of repetitive tasks and the enhancement of human capabilities, generative AI also helps less experienced employees take on more demanding tasks—while effectively aiding in upskilling the existing workforce. Additionally, generative AI can optimize workflows, predict maintenance needs, and translate languages, making manufacturing more efficient and inclusive. This not only improves productivity but also creates more attractive, accessible jobs and draws new talent into the industry.

Expand AI implementation step by step

Looking ahead to 2025, the use of artificial intelligence in manufacturing will accelerate, and its applications will extend to a wide range of use cases. To minimize disruptions and ensure successful technology implementation, manufacturers should introduce AI tools gradually and allow the workforce to adapt to changes in operations.

Prerequisites for successful integration are transparent internal communication, comprehensive training, and continuous support. Additionally important are resources that enable employees to troubleshoot and optimize the use of new technologies themselves. It makes sense to start with small pilot projects—allowing manufacturers to incrementally expand AI implementation and learn what truly works for their operations. Decision-makers in the industrial sector can thus harness the benefits of technology in a controlled manner while working to realize their vision for their own company.