Where’s the AI for Engineers? Practical Tools and Applications
In this interview, digitalization expert Stefan Kaufmann provides an overview of the use and possibilities of AI in the AEC industry – both now and in the future.
Artificial intelligence has reached the AEC industry – and it is here to stay. Whether in design support, data analysis, or construction automation, AI-supported tools are already changing the way we plan and build. But which of these are ready for practical use, and which are still a long way off? Stefan Kaufmann, Product Manager for BIM Strategy and New Technologies at ALLPLAN, discusses this topic.
How is AI currently being used in the AEC industry? What are some of the most important applications, both today and in the future?
Stefan Kaufmann: AI in the AEC industry has developed in two directions: first, there are broadly applicable, universal AI applications such as chatbots and image generators. Secondly, there are highly focused tools that solve specific tasks, such as classifying point clouds or monitoring the energy consumption of buildings. AI is most impressive where broad knowledge is required or where repetitive, low-value-added tasks can be replaced. This frees up time for professionals to engage in strategic and creative activities.
On a practical level, AI is currently used in many areas, such as text translation, researching building standards, and managing project documents. There are also "Any2BIM services" that convert data from drawings or point clouds into BIM models, reducing the manual effort required by planning teams during project preparation.
Pre-trained models now extract structured information from unstructured sources such as PDF plans and use it to create knowledge graphs and intelligently link project data. We are continuously testing multimodal AI models that will support our customers in complex construction projects in the future. The processing of large amounts of 2D information can be significantly optimized with such technologies.
What about generative AI – how does it help in the early stages of design?
Diffusion models, for example, are proving useful for developing initial design ideas. Tools such as Nemetschek's AI Visualizer create images that allow architects to visualize styles and materials in seconds. Modern image generators enable more precise customization, such as the targeted modification of facade materials in image areas or the insertion of people.
There are also initial tools that convert 2D inputs into 3D models. For example, structured BIM data can be generated from simple descriptions via IFC prompts – albeit with limited architectural and structural quality so far. AI can also help with the creation of room books and the understanding of norms and standards, thereby improving design quality in the early stages.
How is AI used in conjunction with BIM?
AI is increasingly becoming a central component of BIM workflows. Analysis tools can be used to query BIM databases directly, for example by asking the model to extract quantities or locate faulty elements. AI also helps to map internal BIM standards to project-specific requirements. This process is currently still manual, laborious, and prone to errors. Another area is model enrichment, as well as the automation of plan creation using the model, which, paradoxically, remains an extremely time-consuming task in BIM projects.
AI is often discussed in the context of sustainability. How can it support greener construction practices?
Implementing sustainable designs is more tedious than complex. AI can make many of these tasks more efficient – for example, assigning material systems to components or recommending suitable materials and design solutions. At the Georg Nemetschek Institute, we are also researching how AI can support the AEC industry on its path to a circular economy. AI's unique ability to recognize patterns in sensor data can help assess the structural integrity of concrete elements without destructive testing, which is particularly relevant for monitoring corrosion of steel reinforcement.
What ethical and legal considerations should the industry be aware of?
Data protection and intellectual property are key issues when it comes to AI. In Europe in particular, clear regulations ensure that data is handled responsibly. At ALLPLAN, for example, we are committed to not using customer data to train AI models and to protecting it in accordance with strict European law when using our AI services. Trust and transparency are crucial. As AI becomes increasingly integrated into everyday work, clear ethical standards are needed to build long-term acceptance and trust.
What AI trends can the AEC industry expect in 2026?
We will see widespread integration of so-called AI agents in all innovative software solutions. AI agents are systems that are capable of pursuing goals autonomously. Embedded in BIM software, they can, for example, record what we as planners are planning to do, develop a solution based on the available program functions, and operate the software for us.
One of the most important trends we can expect in 2026 will therefore be the leap to predictive AI, which will suggest solutions to increasingly complex problems. AI tools will not only be able to generate attractive images, but also assess the consequences of a design decision. Design variants will be evaluated according to qualities such as material efficiency, costs, climate footprint, and energy considerations. AI will gradually evolve from a creative tool to a strategic planning partner for the construction industry.
At the same time, we will see the first signs of data-centric engineering. In terms of construction, this requires well-structured BIM and analysis models, standardized libraries of as-built components, and sensor, operating, inspection, and maintenance data. Companies that systematically build up and feed back their data assets will have a clear advantage when it comes to using AI.
In construction, AI will increasingly find its way from dashboards into robotics and semi-autonomous machines. This affects numerous areas of activity – from earthworks to construction staking, reinforcement work, and masonry construction. Autonomous construction processes will make an increasingly important contribution to ensuring occupational safety, productivity, quality, and overcoming the shortage of skilled workers in the AEC industry.
What advice would you give to companies that are just starting to explore AI?
The most important step is to organize and consolidate your own data. Although generic AI tools are developing rapidly, their effectiveness in the future will depend crucially on the quality and structure of your own data. Make sure your project data is accessible, consistent, and under your control. Only then will you be ready to fully exploit the potential of new tools and innovations.
And finally, what excites you most about AI in the AEC industry?
We are currently experiencing the most significant and comprehensive technological development in human history – at an unprecedented pace. Every week, there are new breakthroughs in the development of AI that would have been unthinkable just a few months ago. The possibilities for solving complex problems with AI are developing exponentially. If we want to fundamentally change the way we plan and build, we must make AI reliably usable in projects and enable professional control. We can only shape the future of the AEC industry with AI together.
