AI in mechanical engineering: from the pilot to practice

Where does mechanical engineering stand from the VDMA's perspective in the productive use of AI - beyond individual lighthouse projects?

Guido Reimann: Mechanical and plant engineering is in a phase of growing relevance and strategic importance of AI. The VDMA survey on Digital Transformation in mechanical and plant engineering from April 2026 shows that more than 80 percent of mechanical engineering companies already attribute greater importance to AI technologies in their own company. Around 31 percent of the companies even use AI solutions productively. This means we are no longer talking only about individual lighthouse projects in the industry. In addition, another 37 percent of companies are currently in AI pilot projects and another 18 percent still want to launch corresponding AI projects by the end of 2026. The topic has arrived in the industry. Pioneers in the current productive operational application areas are software development and development & design as well as corporate management, IT and marketing & communication.

But AI solutions are also being used increasingly in sales and in product-accompanying services for the customers of mechanical engineering. So far, AI has been used less intensively in the areas of finance & controlling, logistics, or legal & compliance, which is also related to the currently lower degree of digitalization in these fields of application. Accordingly, there is still potential here. However, if the initial digital data are missing, then an AI solution cannot help either. To make use of this potential, companies must continue to advance digitalization in these areas as well.

What does the breadth of the industry need most urgently now so that AI moves from pilot projects to series application?

Reimann: First and foremost, and many companies in the industry have already implemented this, there must be a digitalization strategy for the company. Around 60 percent of companies in mechanical and plant engineering already have a digitalization strategy, as the VDMA survey from April 2026 also shows, and a further 19 percent are planning implementation later this year. 

The strategic decision and implementation of broad-based and sustainable digitalization in the company is a basic prerequisite for the broad application of AI technologies and other digital technologies. Also ranked very highly, however, are topics such as change management, speed of implementation, and lack of human resources when it comes to the hurdles in digital transformation in the company. These must also be taken into account and solutions for them worked out.

Because in many cases, it is not due to the purely technical implementation, but rather to organizational challenges, decision-making structures, lack of involvement of those concerned, insufficient knowledge about possible applications and limits, or other non-technical factors, that projects are drawn out or are not transferred into productive operation. 

Particularly in AI projects, however, there is also repeatedly a factor that must be taken into account as well: Not every AI pilot project is destined for success. It is important, with a view to a balanced cost-benefit ratio, to make a decision in good time when no further progress is possible, so that the limited financial and personnel capacities can be used in a targeted manner for other digitalization activities.

What role must an association like the VDMA play so that AI in mechanical engineering scales faster, more safely, and more practically?

Reimann: As an industrial association, the VDMA plays a key role in supporting and accelerating the AI transformation in mechanical engineering. In political advocacy, we are committed above all to ensuring that the framework conditions in Germany and the EU, with regard to international competition, are designed in an innovation-friendly and SME-friendly way as well as in a practical manner for the industry. Furthermore, with a view in particular to better interoperability and availability of industrial data, we also support the development and dissemination of open standards such as OPC UA and the establishment of trustworthy data spaces.

To this end, we participate, among other things, in initiatives such as Manufacturing-X. In addition, for many years we have been promoting the early exchange between research, startups, technology pioneers, IT companies, as well as companies in automation technology and mechanical engineering. Rapid transfer of experience and knowledge in established networks is becoming ever more important in the course of the speed of development of AI technologies in order to advance the building of expertise within one's own company and to quickly find suitable cooperation partners. Through various formats such as studies, white papers, collections of practical examples, and a wide range of event and continuing education formats, together with the member companies we spread this knowledge broadly and provide the impetus for the next AI projects in mechanical engineering and industry.