At the Building 2030 Summer Seminar, doctoral researchers Tuomas Valkonen<\/a> and Roope Nyqvist<\/a> from Aalto University shared fresh insights into how Large Language Models (LLMs) like ChatGPT and DeepSeek perform in construction-related tasks.<\/p>\n The results highlight a simple truth: used wisely, LLMs can accelerate knowledge work; used unwisely, they can produce nonsensical results in areas where precision is critical.<\/p>\n Tuomas Valkonen shared his experiences using ChatGPT without any customization as a helper for MEP designers.<\/p>\n First, Tuomas prompted GPT-4o and DeepSeek to do an HVAC systems description: \u201cI am an HVAC designer, and I\u2019m designing a residential apartment building. The building is heated with district heating and hydronic underfloor heating, with apartment-specific ventilation units. Prepare a systems description of all HVAC systems for me.\u201d<\/p>\n Both LLMs created a \u201cmoderately reasonable\u201d systems description, but DeepSeek missed two essential systems in its output.<\/p>\n Next, Tuomas explained how he provided ChatGPT and DeepSeek with a simplified hospital floor plan showing patient rooms. He asked the LLMs to create a ventilation zoning table for the plan without giving additional instructions about the systems.<\/p>\n Both software programs offered a version with a different number of zones and air volumes.<\/p>\n Tuomas also asked the LLMs to create an HVAC systems schema on top of the floor plan, and ChatGPT to color-code the service zones of an actual plan from a real project. The results were nonsensical. DeepSeek even instructed you to use Microsoft Paint to do the coloring yourself!<\/p>\n Trying to create a room schedule from the plan drawing did not go well either. The plan had 72 rooms, totaling 832 square meters. ChatGPT found 30 rooms covering 313 sqm.<\/p>\n Furthermore, calculating the room-specific heating load or calculating quantities resulted in several errors. ChatGPT failed to correctly perform a simple multiplication on one line, and it suggested that a duct segment in one apartment was two kilometers long.<\/p>\n Eventually, after several iterations, Tuomas started getting better results. He believes that with proper instructions, a quantity takeoff with this technology would become feasible and more reliable.<\/p>\n Tuomas also tested how well ChatGPT could check whether an MEP IFC file included the mandatory information.<\/p>\n Initially, ChatGPT could not read the IFC as-is, but created a piece of software that turned it into a text file it could use.<\/p>\n After a few trials and errors, he discovered a method to obtain the correct results with a model having 5,000 components. A model with 50,000 components, however, proved to be too large. In that case, ChatGPT could create Python code to do the job.<\/p>\n As Tuomas\u2019s tests demonstrate, out-of-the-box LLMs can\u2019t be reliable assistants in many common construction tasks that deal with graphical information and understanding of construction concepts. However, they shine in some other applications.<\/p>\n Roope Nyqvist discussed use cases where LLMs proved helpful. He had, for example, developed a custom GPT that can answer questions about hospital design. The knowledge base<\/a> incorporates 6,543 files, covering 1,749 pages of expert knowledge, into a system available for anyone to use. It took only 100 hours to create the tool.<\/p>\n LLMs also excel at creating clear RFIs, change orders, meeting summaries, or client-friendly reports. They can convert complex technical content into straightforward language or translate between languages for international project teams.<\/p>\n In bids and proposals, LLMs accelerate narrative writing, case study creation, and market research. They are already proving useful in helping contractors tell their story more effectively and quickly.<\/p>\n I asked ChatGPT 5 to summarize the reliable and unreliable use cases in the construction industry. It provided me with a list that is visualized in the following diagram.<\/p>\n It\u2019s clear that domain-specific models are essential when accuracy and reliability are necessary. LLMs won\u2019t replace Togal, Buildots, Kreo, and similar tools, but they will communicate with them.<\/p>\n ChatGPT is most reliable in language-heavy, reasoning-heavy, low-liability tasks: communication, documentation, learning, project support, and orchestration. And even then, ask it: \u201cAre you sure?\u201d<\/p>\n PS. You can watch Tuomas\u2019s and Roope\u2019s presentations in Finnish <\/em>on YouTube<\/em><\/a>.<\/em><\/p>\n<\/div>\nHVAC systems descriptions with an LLM<\/h2>\n
Using PDFs as input for system design<\/h2>\n
Quantity takeoffs, way off<\/h2>\n
Assessing IFC file data quality<\/h2>\n
Where do LLMs shine?<\/h2>\n
![\"LLM](\"https:\/\/aec-business.com\/wp-content\/uploads\/2025\/09\/use-cases-llm.png\")
Specialized tools rule<\/h2>\n