In our previous article, we examined how generic artificial intelligence solutions often fail in the healthcare context. Today we explore how this lesson applies to the construction industry, an equally complex field that requires specialized solutions.

‍

Introduction: beyond generic AI

Artificial intelligence has captured the attention of business leaders in every industry. However, as recent experiences in the healthcare and construction industries show, generic AI solutions often fail when applied to highly specialized domains. True transformation comes not from applying general capabilities to specific problems, but from building AI that fundamentally understands the domain from its foundation.

‍

This truth clearly emerges from an analysis of the construction and real estate industry, where multidisciplinary complexity, market fragmentation and stringent regulations create unique challenges that only specialized solutions can effectively address.

‍

The need for specialized sectoral understanding

‍

Terminological and normative misalignment

Generic AI models fail to properly distinguish between fundamental technical concepts such as "load-bearing walls" and "partitions," or between "slab foundations" and "pier foundations," leading to misinterpretations in projects where accuracy is vital for safety. This terminological mismatch also extends to regional variations: an Italian "brick slab" has different characteristics from a northern European slab system, with critical implications for structural and earthquake-resistant calculations.

‍

Similarly, the strict regulatory framework of the construction industry, with building codes, safety standards, and environmental regulations that vary by region, presents a challenge that generalist IA solutions rarely can address. The Eurocodes and the Italian Technical Standards for Construction (NTC) have substantial differences in safety coefficients that a generalist IA cannot discriminate, with potentially serious consequences for structural safety.

The recognition of transformative potential

‍

Despite these challenges, the potential of AI in transforming the industry is widely recognized. According to JLL's 2023 Global Real Estate Technology Survey, AI and generative AI were ranked among the top three technologies that will have the greatest impact on the real estate industry in the next three years by investors, developers, and corporate occupiers. However, the same respondents indicated the least understanding of AI compared to other technologies such as blockchain, virtual reality, and robotics.

‍

This apparent contradiction underscores the need for specialized approaches that can bridge the gap between recognized potential and actual implementation.

‍

The specialized approach: success stories in the construction industry

Specialized building AI solutions are already proving their value through concrete case studies:

‍

Reduction of design errors

In a large residential project, the implementation of an industry-specific intelligence module led to:

• 68% reduction in design errors
• 23% decrease in review time
• Estimated 15% savings on total costs
• Significant improvement in delivery time

Particularly notable was the impact on the management of in-process variances, historically responsible for cost increases of up to 20-30%. The specialized platform reduced these impacts to 7%, thanks to the ability to automatically propagate changes to all related project documents.

‍

Optimized materials management

An infrastructure developer implemented a specialized materials management module that resulted in:

• 31% reduction in inventories
• 24% decrease in delivery delays
• Savings of more than 2 million on logistics costs
• Improved sustainability with less material waste

A crucial but often overlooked aspect was the impact on cash flow management. Procurement optimization reduced fixed assets by 42 percent, significantly improving the company's financial position.

‍

Optimization of site scheduling

A construction company specializing in complex urban redevelopment implemented space-time optimization algorithms that resulted in:

• Reduction of 28% in overall processing time
• 34% increase in the efficiency of work teams
• Decrease of 62% in interference between simultaneous workings
• Improved predictability of timing, with final deviations of less than 5%

This case highlights how specialized AI can solve one of the industry's endemic problems: the difficulty of scheduling in complex contexts with multiple variables and constraints. Traditional project management techniques such as CPM or PERT show significant limitations in real-world scenarios, while the AI-based approach has demonstrated measurable operational superiority.

‍

The broader view: transformation of the housing market

The impact of AI extends beyond construction, transforming the entire real estate industry across five key dimensions:

‍

1. Geolocation and clustering

Companies and investments in AI tend to be concentrated in established technology markets. JLL's research shows accelerated demand for AI talent, with job postings increasing by more than 250 percent since the beginning of 2021. In the long term, this growth is likely to be concentrated where AI talent is available: established primary and secondary technology hubs, innovation centers, and universities.

In the United States, 42 percent of AI companies are concentrated in the San Francisco Bay Area, followed by Boston, Seattle and New York, with a projected real estate growth of 1.6 million square meters by the end of the year in the United States alone.

‍

2. Alteration of demand among assets

The development of AI requires more and better data centers, energy networks and connectivity infrastructure. According to the JLL Global Data Center Outlook 2023, the global colocation data center market is expected to grow 11.3 percent annually from 2021 to 2026, while the hyperscale data center market is expected to grow even faster, at about 20 percent annually.

AI's infrastructure location criteria give greater weight to lower energy prices and lower land costs, driving growth to less crowded markets such as Atlanta in the United States, Malaysia and Thailand.

‍

3. New types of assets and products

The emergence of the "truly intelligent building" is imminent. AI-compatible infrastructure will become a default standard, just as internet connections are a default feature of today's buildings. AI will also help realize zero-emission buildings with high sustainability performance.

This aligns with the "dynamic digital twins" described in the building industry, which move beyond the static concept of BIM toward models that evolve in real time throughout the building lifecycle, enabling predictive maintenance management that reduces operating costs by 23-31% and increases the useful life of facilities by 15-20%.

‍

4. New investment and revenue models

AI-enhanced underwriting and processes will enable faster transactions and more efficient understanding of properties and markets, catalyzing investment on a global scale. AI-enabled infrastructure and the ability to connect multiple systems could also enable the expansion of "space as a service" models and new revenue streams for owners and developers.

A concrete example cited in the JLL report is that of Royal London Asset Management, which saw significant improvements in HVAC operations and energy efficiency in an 11,600-square-foot commercial building. By implementing JLL's AI technologies, the company achieved a record ROI of 708 percent and energy savings of 59 percent, reducing carbon emissions by up to 500 metric tons per year.

‍

5. New approaches to the design and functionality of spaces

AI will enable experience-driven design and highly customizable environmental settings. This complements the multimodal AI for inspection described in the construction industry, which will combine understanding of text, images, and data from drones and IoT sensors to monitor the progress and quality of construction, with particular promise in integration with LiDAR technology for real-time structural monitoring.

‍

‍

The socio-economic dimension: impact on work and skills

Contrary to fears of substitution, the data collected show that specialized AI is having a positive impact on the workforce:

Enhancement of existing skills

Specialized AI has enhanced the role of skilled craftsmen, freeing them from administrative tasks and allowing them to focus on the quality aspects of workmanship. This has led to an increase in perceived quality and a reassessment of technical skills.

‍

This approach aligns with Microsoft CEO Satya Nadella's vision that AI service providers are making the conscious choice to explore a human-centered approach, developing "co-pilot" products designed to assist people, rather than "autopilot" products that aim to completely replace human roles.

‍

Transformation of job profiles

New hybrid roles are emerging, such as the "BIM Construction Manager" and the "Digital Construction Specialist," with skills straddling traditional construction and digital technologies. These profiles command salaries 35-40% higher than the industry average.

According to Goldman Sachs, which cites a study by MIT economist David Autor, more than 85 percent of U.S. job growth over the past 80 years is explained by the creation of new positions driven by technology.

‍

Democratization of experience

AI's ability to codify and make accessible best practices has narrowed the performance gap between small and large firms, promoting fairer competition based on actual quality rather than firm size.

‍

The future: emerging innovations and strategic approach

Impending technological advances

In the construction industry, future innovations include:

• Predictive analytics for site safety: Models that preemptively identify risk situations based on historical data and site configurations, with a 76% accident prediction capability and a potential 58% reduction in serious injuries.
• Multimodal AI for Inspection: Features that integrate understanding of text, images, and data from drones and IoT sensors to monitor construction progress and quality.
• Integration with site robotics: Early pilot projects with floor-laying robots and automated finishing systems have shown productivity gains of up to 300% in repetitive operations, with superior quality and reduced scrap.

In the broader real estate sector, JLL highlights that the enterprise use case market for generative AI is expected to reach $42.6 billion in 2023, growing 32 percent annually to $98.1 billion by 2026.

‍

Strategic and responsible adoption

Organizations must consider how to harness the power of AI to support their business goals in a responsible and ethical manner. JLL stresses the importance of being vigilant about three types of emerging regulations:

1. Market standards and protocols related to data quality, intellectual property rights, privacy and data security.
2. Regulations to mitigate social risks, such as measures to protect the labor market from shocks or safety standards for autonomous vehicles.
3. Environmental legislation, particularly that aimed at mitigating carbon emissions from the growing digital economy.

Organizations will need to think through a number of key questions: What does the growth of AI mean for investment and localization strategies? What existing or future applications of AI need to be prepared and tested now? What are the potential business and social risks?

‍

Conclusion: the value of the specialized approach

As in healthcare, true transformation in construction and real estate does not come from applying generic AI to complex problems, but from solutions built specifically for the unique challenges of the industry.

‍

Construction is an emblematic case of a high-complexity, low-digitization sector: it is second to last among industries in terms of digital adoption rate. These very characteristics make it an ideal terrain for demonstrating the value of specialized AI over generic solutions.

‍

The peculiarity of the construction sector lies in its being simultaneously knowledge-intensive and labor-intensive, with a delicate balance between cognitive and operational dimensions. This dualism requires AI systems that are not limited to data processing, but deeply understand the decision-making and operational processes that characterize the industry.

‍

As one project manager of a major architectural firm observed, "The difference between general and specialized IA in construction is like that between a general laborer and a specialized master. Both have value, but when it comes to complex projects, specialized expertise becomes indispensable."

‍

The challenge for the future will be to strike the right balance between vertical specialization and horizontal interoperability, enabling different players in the supply chain to benefit from tailored solutions that can still talk to each other. Only in this way can AI fulfill its promise of transforming one of the most innovation-resistant sectors into an example of efficiency, sustainability and quality.