Only 11% of construction and development executives in a new BRG construction AI survey say they’ve cut staff because of artificial intelligence, while 18% say it led them to add people in certain areas. The consulting firm released the report, AI in Construction: Measurable Results, Unclear Benchmarks, on Oct. 14.
The headcount story is smaller than the spending story. BRG says 68% of respondents have spent at least $250,000 on AI over the past 24 months and 37% have spent $1 million or more. Just 15% say AI is fully integrated across projects and operations.
How AI is changing construction staffing
The workforce effects BRG found show up in structure, not payroll. More than half of respondents changed how much they rely on outside vendors, consultants or contractors. Another 46% changed team structure and collaboration, and 38% reskilled or retrained existing employees, according to the firm’s release.
That outside-vendor number is the one consultants and specialty firms should read twice. BRG doesn’t say which direction it moved. An owner that brings early estimating in house with an AI tool needs fewer outside estimators. One that buys AI expertise it can’t hire needs more consultants. Both count as “changed reliance.”
Adoption is broad. Respondents reported AI use in preconstruction (46%), construction (49%), post-construction (40%) and business operations (43%). Planning and feasibility is the most common single function at 58%, rising to 80% among builders. Dara Hendrix, a BRG managing director, said that’s where much of the project risk sits and where early estimating work generates little direct revenue.
The skills gap behind construction AI adoption
The top obstacle is people. Some 40% cite a lack of skilled talent or expertise as likely to limit AI’s benefits, ahead of limited relevance to their work at 38% and data and integration problems at 37%. So the same firms that aren’t cutting headcount say they can’t find the people to run the tools.
That fits other recent data. BLS-based analysis this month found AI exposure is low in 45 of 47 construction occupations, with managers and inspectors the exceptions. Office roles change first. The field barely notices.
The risk side hasn’t caught up either. Liability for errors in AI-generated work worries 43% of respondents and unclear contract treatment worries 40%, but only 42% say their policies cover human oversight and accountability and just 34% cover accuracy and reliability.
What the BRG numbers can and can’t support
Read the sample before the headlines. BRG surveyed a bit more than 100 executives in July 2026, mostly in North America, and mixed builders with developers. Responses are self-reported. Double-digit cost savings were widespread, BRG says, with 67% reporting them during construction, but 6% don’t measure ROI at all and only 37% rely mainly on financial metrics.
BRG also sells AI advisory work, and the report is the fourth in a series it runs across industries. Travis Olson, a BRG director, argued that a narrow ROI focus can miss work AI makes possible for the first time. Maybe. It also makes “measurable results” harder to measure.
For anyone planning next year’s staffing, the takeaway is plain. AI budgets are buying new roles, retraining and different vendor mixes. They aren’t buying fewer people, at least not yet.