At a Glance

  • A recent report by McKinsey Global Consulting indicates that AI is not an "extinction event" for AEC (Architecture, Engineering, and Construction) firms, but it will fundamentallychange the way construction professionals optimize specific tasks。
  • The report, titled "How AI is Reshaping the Future of the AEC Industry," released on July 15, describes two camps of AI users: one that leverages AI to automate core tasks, and another that uses it merely as a superficial productivity tool. The report specifically notes that firms that control proprietary project data, decision-making workflows, and can charge based on outcomes rather than processes will benefit the most.
  • The report states: "Early adopters report productivity gains in design, modeling, and construction feasibility workflows, but these advantages are likely to quickly become table stakes for the industry."

Deep Insights

Overall, McKinsey advises construction firms to use AI to transform "domains" (i.e., end-to-end processes). These processes can be independently redesigned due to their smaller scale while still having a substantial impact on the business, rather than deploying isolated point solutions.

Daniel Ahmoye, a partner in McKinsey's Calgary office, said in an interview: "As models become more prevalent, I think the advantage will come from firms that can most quickly redesign how they work, their roles, their operating models, and how they think about business models."

McKinsey's research claims that AI has the potential to automate 39% of non-manual work in the construction industry, and in architecture and engineering, this figure could reach 50%.

Overall, McKinsey identified 150 workflows across 25 AEC-related domains, which vary in their potential for AI and automation application. The report specifically notes that skills such as data entry, invoice processing, and equipment inspection will undergo the most significant changes by 2030.

McKinsey categorizes these automation targets into three time phases:

  • Near-term (first 18 months): Focus on streamlining end-to-end workflows, including functions such as bid/no-bid analysis, cost estimation, and proposal drafting.
  • Mid-term (18 months to 4 years): Contractors leverage data advantages to automate handling of proprietary data, such as requests for information (RFIs), drawings, specifications, and as-built reports.
  • Long-term (4+ years): Construction firms can deploy AI on job sites, including the use of autonomous construction equipment and coordinating transportation scheduling between factories, yards, and sites.

However, Ahmoye points out that these timeframes do not mean certain tasks will completely disappear. By 2030, only some tasks will be automatable, and when considering the role of people in task completion, the scope of automation narrows further.

Ahmoye said: "Only certain roles can be automated, and it's more about the tasks and activities themselves."

Although the numbers seem large and daunting, what's actually involved are fragmented pieces within a broader set of activities.

Ahmoye added: "What really matters is how these activities are linked together, combined in ways that reduce friction, and create smoother workflows throughout the entire construction project lifecycle."

McKinsey has previously engaged in the construction productivity debate, most notably with its2017 report on the construction industry's failure to keep pace globally. Another report shows that from 2000 to 2022,global construction productivity increased by only 10%, as noted by McKinsey experts in an article on Construction Dive.

The report also issues a caution regarding the ongoing debate among construction firms between "building their own solutions" and "buying technology from software developers." Notable construction firms such asSuffolk ConstructionandTurner Constructionhave entered this space, developing internal tools to address pain points.

The report notes: "AEC firms have historically struggled to build and scale software products, and AI is advancing too rapidly for most incumbents to rely primarily on internal development." Instead, construction firms should focus on their own advantages that are difficult for competitors or suppliers to replicate, such as proprietary data or customer relationships.

To this end, firms should build in areas where their own expertise is the product; and buy in areas where external vendors invest far more than the firm could ever afford. Ahmoye said: "So, for me, it's not a buy-or-build question, but rather where do we get our advantage from."

Looking ahead, McKinsey highlights several steps to help construction firms prepare for AI integration, including: prioritizing 3 to 5 high-value workflows; deciding where to buy, build, or partner; scaling through governance mechanisms; and measuring what truly matters.