Although the potential of Artificial Intelligence (AI) in the construction and real estate industry is recognised, current industry figures show a reluctance to make corresponding investments. This development brings the question of the reasons for this hesitant attitude into focus ahead of EXPO REAL. The debate concentrates not primarily on the functionality of the technology, but on the quality of the data with which it operates.
A large-scale study conducted by PlanRadar among 1,728 construction and real estate professionals in 14 countries illustrates this situation. 58 per cent of respondents attest to AI's ability to significantly mitigate daily challenges, particularly adherence to schedules and dealing with project changes. Nevertheless, almost half of companies do not plan to invest in AI-powered tools. The analysis shows that this reluctance is not due to fear of job losses; only 6 per cent of respondents expressed this concern. This correlates with a persistent shortage of skilled workers in the industry, where AI is perceived as a solution to alleviate this bottleneck.
Data Quality as a Central Challenge
The primary hurdles lie instead in the accuracy and reliability of AI results and in data protection. This illustrates the fundamental connection that AI's performance directly depends on the quality of the data processed. Many companies do not yet have the necessary structures for consistent data collection, which are indispensable for reliable AI applications. For project developers and clients who commit capital and bear project risks, well-founded answers to questions about project status, potential risks, and open decisions are crucially important. This involves increased transparency, risk minimisation, and the basis for sound decisions.
The transformation of AI from a mere reference tool to an active assistant is a decisive factor here. AI agents can not only collate information but also prepare tasks, analyse project data, initiate processes, and transmit structured results to relevant stakeholders. An example of this is an AI agent that continuously monitors a project, converts on-site observations into structured tickets via voice, and assigns these to the appropriate process. This agent can also summarise project progress and highlight risks early on to support decision-making processes. Expert approval remains with humans, while preliminary and follow-up work runs automatically in the background.
Prerequisites for the Reliable Use of AI
Effective AI deployment requires a fundamental review of existing processes. If AI systems are introduced without questioning existing workflows, a significant part of the potential benefit is lost, and there is a risk of receiving only seemingly convincing results based on incomplete data. Martin Pietzonka, Senior Head of Innovation at the Drees & Sommer Group and Head of Startup Hub, emphasises that AI represents a fundamental change in value creation from data and that its performance directly depends on the quality of information.
- —First-class data capture from the outset, to ensure clear assignment of photos, plans, tickets, and protocols to projects.
- —Process revision instead of mere digitisation, to adapt workflows to the real-time capabilities of systems and clearly define responsibilities.
- —Seamless traceability of AI recommendations and automated processes, especially for clients and insurers in liability-relevant contexts. PlanRadar relies on certified ISO security standards and the exclusive use of customer data for the respective project context for this purpose.
Structured and accessible project data are thus not a technical triviality, but the fundamental prerequisite for reliable AI deployment in practice. The actual added value of AI is not manifested in demonstrative applications, but in its ability to rapidly convert project data into traceable decisions. The technology is already available in many companies; the challenge lies in creating the necessary connective infrastructure.
