By Neil Sahota
If you believe the headlines, construction is undergoing an AI renaissance. Robotics. Digital twins. Smart cities. Yet, when I walk job sites or sit with construction executives, the people actually delivering projects (electricians, plumbers, concrete crews, HVAC teams) are still buried in PDFs, spreadsheets, and phone calls.
This disconnect explains why construction productivity has barely improved in 50 years, even as costs and complexity soar. What I’ve learned? The real AI opportunity in construction isn’t megaprojects. It’s subcontractors.
Where Margins Go to Die
Subcontractors generate the majority of construction value but capture the least margin. Why?
- Manual estimating under extreme time pressure
- Change orders discovered too late to recover cost
- Poor visibility into labor utilization and schedule risk
- Compliance and documentation overload
In multiple engagements I’ve advised, subcontractors lost 3–7% of project margin (not because of bad execution) because they didn’t see problems early enough to act.
AI changes that.
Estimating: The Quiet Revolution
Most subcontractors still estimate based on drawings using manual takeoffs. AI-assisted estimation can flag scope gaps, historical underbids, and material volatility in minutes. In one case, AI review of historical bids identified a pattern of systematic underpricing on change-heavy scopes, something human estimators felt but couldn’t prove. This insight alone reshaped bidding strategy within a quarter.
Don’t get me wrong, though. This isn’t about replacing estimators. Instead, it’s about giving them memory, pattern recognition, and foresight.
Safety: The Conversation No One Wants
Construction safety analytics often focus on general contractors; however, most incidents occur at the trade level. AI visual recognition systems trained on real job-site behavior (not the idealized safety manuals) identified precursors to incidents days before they happen. I’ve seen pilots reduce recordable incidents without issuing a single new rule. Why? AI doesn’t lecture workers, but it does notice the risk humans normalize.
Talent: The Skilled Labor Cliff
The construction industry faces a severe labor shortage, with an estimated 500,000+ open roles in the U.S. alone. As a result, many constructions turn to AI to capture tribal knowledge, standardize workflows, and assist onboarding because they are becoming existential, not optional.
However, subcontractors don’t have innovation budgets. They have payroll. As a result, the industry suffers from a deadlock problem. Most AI vendors chase enterprise buyers, not fragmented trades. This leaves enormous value untouched in construction and creates a strategic opening for leaders willing to move first.
Ultimately, the future of construction AI won’t be decided in glossy demos. Rather, it will be decided in trailer offices, at 5:30 a.m., by people who need results… not buzzwords.
About the author:
Neil Sahota is an IBM Master Inventor, United Nations (UN) Artificial Intelligence (AI) Advisor, AI Strategist, and author of two books, Own the A.I. Revolution and AI Activation Code. With 20+ years of business experience, he works with organizations to create next-generation products/solutions powered by emerging technology. His work experience spans multiple industries, including legal services, healthcare, life sciences, retail, travel and transportation, energy and utilities, automotive, telecommunications, media/communication, and government.
