Data and cost barriers blocking AI’s sustainable construction potential

Artificial intelligence (AI) is transforming the way buildings are designed, constructed, and maintained, but critical challenges remain in scaling these technologies for sustainability. A new study provides the most comprehensive assessment yet of how digital tools combined with AI are reshaping the built environment.

The paper, titled AI-Powered Advanced Technologies for a Sustainable Built Environment: A Systematic Review on Emerging Challenges, and published in Sustainability systematically reviews more than 100 studies published between 2015 and 2025, mapping how technologies such as Building Information Modeling (BIM), the Internet of Things (IoT), Digital Twins, machine learning, and optimization techniques are being applied across the building life cycle. The findings point to significant progress but also highlight blind spots that must be addressed for AI-driven sustainability to reach its full potential.

Where AI is driving the most impact

The review shows that the integration of AI with BIM, IoT, Digital Twins, and optimization methods is having the strongest effect in two key stages: building design and operation. In design, BIM combined with AI supports sustainability assessments, low-carbon material choices, and circular economy planning. Machine learning models are increasingly used to simulate design alternatives and reduce environmental footprints before construction begins.

In operation and maintenance, which accounts for the majority of research, IoT sensors and Digital Twins enable real-time monitoring of building systems. When coupled with AI, these technologies drive energy optimization, predictive maintenance, and occupant comfort strategies. Reinforcement learning and advanced neural networks are being applied to manage heating, cooling, and photovoltaic performance, while optimization algorithms translate these insights into actionable decisions.

The study emphasizes that these technologies are already producing measurable gains in energy efficiency, emissions reductions, and cost savings. Yet their adoption remains uneven across sectors and phases of the building life cycle.

What are the gaps holding back widespread adoption?

While the benefits are clear, the review identifies several persistent barriers. Interoperability remains one of the most serious issues, with fragmented platforms preventing seamless integration of BIM, IoT, Digital Twins, and AI. Without standards that allow data to flow smoothly between systems, the potential of these technologies is undermined.

Data quality and access are another critical limitation. Many models suffer from poor generalizability because datasets are fragmented, proprietary, or insufficiently representative. This hinders reproducibility and reduces the trust of stakeholders. Cost is also a barrier, with many AI-powered solutions still too expensive for widespread deployment, particularly in resource-constrained regions.

The study highlights a lack of research attention to the renovation and end-of-life phases of buildings, as well as to sustainable heritage preservation. While design and operations dominate the literature, these neglected areas are essential for advancing circular economy goals and reducing waste. Geographic disparities also persist, with most research concentrated in Europe, China, India, Australia, and the United States. Latin America and Africa remain underrepresented, reflecting broader digital infrastructure gaps.

How should policy and research respond?

The authors argue that scaling AI-powered sustainability in the built environment requires coordinated action from researchers, policymakers, and industry. They call for greater investment in open datasets and interoperable standards to reduce fragmentation. Public datasets, in particular, would allow models to be tested, validated, and replicated across contexts.

Research must also expand beyond operations and design to address renovation, end-of-life, and heritage preservation. These phases are critical to achieving circular economy objectives but remain underexplored. Policymakers should prioritize funding and regulation that encourages digital adoption in these areas.

Capacity building is another pressing issue. The review notes that the technical expertise and infrastructure required to deploy advanced AI models are lacking in many regions. Training programs, knowledge-sharing networks, and investments in cloud computing infrastructure could reduce these disparities and accelerate adoption.

Finally, cost-effectiveness and regulation remain key challenges. The study calls for frameworks that translate sustainability rules into machine-readable formats and financial models that make advanced technologies accessible to smaller firms. Without addressing these systemic issues, the gains made in pilot projects risk remaining confined to elite markets rather than spreading across the global building sector.