As we stride into 2025, the construction industry is abuzz with a new kind of arms race—one that’s driven by artificial intelligence. Industry giants like Balfour Beatty and Skanska are not just embracing AI; they’re developing their own tools to streamline construction processes. Meanwhile, firms like Cork Howard are leveraging third-party solutions such as Billy the Bot, created by Stampli, to manage invoices efficiently, freeing up valuable time and resources. The conversation around AI in construction is shifting from mere potential to practical application, as evidenced by the Engineering News-Record’s FutureTech conference in 2024, where builders discussed AI’s role in contract compliance and enhancing jobsite safety.
Now, a new contender is emerging: agentic AI. This advanced form of AI, touted by tech giant Nvidia as a “knowledge robot,” promises to automate complex tasks across industries. But is it just hype, or does it hold genuine transformative potential for construction?
Agentic AI builds on generative AI but takes a significant leap forward. Unlike traditional AI, which responds to specific queries, agentic AI can act autonomously, tackling multi-step problems through sophisticated reasoning and planning. For instance, while a generative AI bot might provide a one-off answer to a query about subcontractor payments, agentic AI could autonomously create a detailed plan around bills and invoices, presenting a comprehensive report to human overseers.
Barry Mostert, Oracle’s senior director of artificial intelligence and analytics, believes agentic AI could revolutionize construction. AI agents could enhance safety protocols, optimize costs, and proactively mitigate risks. Patrick Murphy, CEO of Togal.AI, envisions agentic AI chatbots functioning like employees, generating and analyzing multiple plans, and proposing new ideas based on market trends.
Yet, not everyone is convinced. Some analysts, like Vladimir Lukic from Boston Consulting Group, caution that while agentic AI holds promise, it also comes with security concerns and potential pitfalls. Senthil Kumar, CTO of Slate Technologies, echoes this sentiment, noting that AI in construction has a history of overpromising and underdelivering. He advises contractors to implement agentic AI strategically, focusing on specific areas like schedule management, risk mitigation, and resource optimization.
Craig Le Clair, principal analyst at Forrester, is even more skeptical. “Most agents are far from agentic. This is the future and does not yet exist,” he asserts. Le Clair’s thorough review of AI agents in production reveals that many fall short of true autonomous behavior.
Kumar also warns of red flags when evaluating AI vendors, such as systems that can’t explain their outputs or guarantee full autonomy. He stresses the importance of human-in-the-loop oversight, ensuring that AI enhances rather than replaces human expertise.
Agentic AI’s potential to reshape the construction sector is undeniable, but it is not a panacea. As the industry navigates this new terrain, the key to success will be a balanced approach—leveraging AI’s strengths while acknowledging its limitations. Contractors who integrate agentic AI thoughtfully, aligning it with their strategic goals, stand to gain a competitive edge.
The future of construction is increasingly digital, and AI is at the forefront of this transformation. Yet, it’s not about putting everything on autopilot; it’s about finding the right balance between human expertise and AI-driven insights. The industry is on the cusp of a significant shift, and how it navigates the integration of agentic AI will shape the development of sustainable, efficient, and innovative construction practices.
