[Digital Today reporter Yoonseo Lee] A data centre boom driven by the spread of artificial intelligence is bringing long-standing problems in the construction industry to the surface.

TechRadar, an IT media outlet, reported on June 9 local time that a shortage of skilled workers, rising materials costs and operational inefficiencies have combined to make digital transformation, especially before construction starts, a key task.

Construction, by nature a site-centred industry, has long relied on manual quantity takeoffs, fragmented spreadsheets and outdated estimating methods. In normal times, that has merely eroded time and margins, but a recent surge in data centre demand is being cited as turning such inefficiency into a bottleneck that blocks the execution of core infrastructure.

Behind this is the expansion of large-scale investment by Big Tech. Amazon Web Services (AWS), Microsoft (MS) and Meta are directly investing hundreds of billions of dollars to expand data centres. The U.S. data centre market is projected to grow by 10 to 11 percent a year on average over the next decade, and as demand for infrastructure supporting cloud computing and AI search rises rapidly, the construction industry has found it difficult to respond with existing methods.

The biggest problem is labour. Associated Builders and Contractors (ABC) estimated that 349,000 new workers will be needed in 2026 to meet construction demand. It also said an additional 456,000 workers will be needed the following year. With labour shortages compounded by rising costs, development projects that are being halted or put on hold in the private sector have increased to about twice last year's level.

In this situation, the solution drawing attention in the industry is not technology that replaces on-site workers, but technology that raises productivity. Contrary to perceptions that AI takes away jobs, digital transformation and AI-based workflow automation may be a way to make the current workforce more efficient, rather than replace people, and help keep up with demand.

The growing importance of the pre-construction stage was also emphasised. If estimating errors emerge after a contract is won, profitability can be shaken sharply. For contractors, the most feared moment is right after winning a bid, and the likelihood of errors rises as well when using long-standing manual methods or older desktop software. Accurate bidding starts with precise measurement, and precise measurement requires technology.

When entering a new industry such as data centres, materials, labour costs and pricing structures may be unfamiliar. In that case, tools are needed to systematically organise unit costs per area, expected profit and detailed work items. It must also be possible to interpret complex drawings and calculate the required quantity of materials accurately down to each nail to protect margins.

Reducing the disconnect between the site and design was also presented as a task for digital transformation. In the past, once design approval was granted, a one-way structure in which site teams built as-is was common, but during actual construction it was not rare for design and on-site practicality to diverge. As a result, a circular approach that feeds site data back into design revisions within an AI-based system is becoming more important.

The core of this trend lies not in the data centre boom itself but in changes to construction processes that can handle it. With labour shortages worsening, digitalisation in the pre-construction stage and AI automation are emerging as ways to raise on-site productivity.