Construction teams are navigating growing project complexity, compressed schedules, increasing documentation requirements and an expanding volume of digital information. At the same time, owners and stakeholders expect greater transparency, stronger coordination and faster decision-making — often without additional resources.
As artificial intelligence (AI) becomes more embedded in the tools project teams rely on every day, the conversation is shifting from experimentation to practical application. At the recent Construction Management Association of America’s (CMAA) Focus26 conference, industry leaders explored a more immediate and pragmatic question: how can AI be used today to reduce administrative burden, improve clarity and support better project outcomes?
In this roundtable, STV’s Jarvis Alridge, CCM, senior project controls manager; Eric Pearson, CCM, project manager; Kristin McElroy, CCM, digital development director; and Mark McElroy, director of digital solutions, discuss how AI can be thoughtfully integrated into existing processes to streamline communication, strengthen coordination and help project teams focus more time on high-value work.
How does your experience in project controls and program delivery shape the way you think about AI in everyday project work?
Jarvis Alridge: Working in project controls means managing large volumes of schedule, cost and performance data while keeping teams aligned and informed. Over time, I’ve seen that the biggest opportunity isn’t adding new tools – it’s reducing friction in the workflows teams already rely on. Built-in AI is most effective when it helps organize information, improve clarity and streamline administrative tasks like drafting summaries, narratives and reports. When AI is embedded directly into familiar platforms, teams can save time without disrupting established processes or replacing professional judgment.
Kristin McElroy: In project controls, consistency and traceability matter just as much as speed. Teams often capture the same information multiple times – in meeting notes, field reports and follow-up emails – which increases risk and creates opportunities for misalignment. AI is most effective when it reduces that duplication by translating raw inputs into clear, structured documentation that aligns with how projects are governed and reported.
Eric Pearson: From a program delivery perspective, the challenge is rarely a lack of information – it’s the effort required to manage it. Overseeing complex capital programs has shown me how much time teams spend keeping project information organized, aligned and accessible. AI becomes valuable when it helps structure that information automatically, so teams can spend less time managing data and more time focused on coordination and decision-making.
Why is the message of “using the tools you already have” resonating with project teams right now?
Jarvis Alridge: The message of using the tools you already have is resonating with teams because it allows teams to stay or become flexible and adaptive while avoiding the barrier of cost. Once many professionals realize that AI is already embedded in the tools they use every day it empowers them to explore ways to harness the immense computing power of AI to optimize their day-to-day tasks. When teams understand what’s already available within familiar tools like PMIS, document creation, presentation and reporting platforms, adoption or creation of more efficient workflows feels achievable. The real shift isn’t about acquiring new technology; it’s about learning how to apply AI intentionally within existing workflows to reduce repetitive work and improve outcomes.
Eric Pearson: Teams are overwhelmed by options and new technologies. The idea of working within familiar tools resonates because it lowers the barrier to adoption. When AI is integrated into workflows teams already use – like email, meetings and reporting – it can start removing administrative friction immediately, without adding complexity or new systems to manage.
Mark McElroy: From a digital solutions standpoint, we want to advise our clients on how to best leverage their existing technology without spending money on new tools. This resonates because teams are overwhelmed with too many systems and are realizing that many already have underutilized AI and automation capabilities. Focusing on optimizing what they have is faster, lower risk and delivers immediate value without added complexity.
What’s a practical AI use case that teams can adopt immediately – without changing platforms or reinventing their process?
Mark McElroy: Email management is a good starting point, since email remains the backbone of project communication, and small efficiencies here quickly scale across the team. With Copilot integrated into Microsoft Outlook, teams can automatically summarize long email threads, draft responses, extract action items and highlight key decisions or risks—without leaving their inbox. I have built an automated weekly project summary email that’s delivered to my inbox, highlighting key updates, identifying upcoming action items and listing due dates, giving me a clear, consolidated view of what’s next without manually pulling information from multiple threads.
Eric Pearson: From a program delivery standpoint, the value isn’t just faster documentation — it’s what that documentation enables next. When meeting and field information is captured accurately and structured consistently, teams can identify risks sooner, keep stakeholders aligned and make decisions with greater confidence. That continuity is especially important on complex programs where small gaps in communication can quickly turn into schedule or cost impacts.
Kristin McElroy: Another practical use case is using Copilot across the Microsoft ecosystem to generate a monthly project status report by pulling from multiple existing sources. Copilot can synthesize data from Power BI dashboards, Excel cost trackers and even prior PowerPoint updates to draft a cohesive narrative on budget, schedule, risks and key milestones.
Instead of manually compiling inputs, a project manager can prompt Copilot to “create this month’s executive summary,” and it will generate a first draft in PowerPoint or Word, complete with visuals, summarized trends and key action items, saving hours while improving consistency and data alignment.
What types of project tasks are best suited for AI support today?
Jarvis Alridge: Project teams spend a significant amount of time on administrative and coordination tasks, drafting meeting summaries, organizing information, tracking updates, mapping trends, producing forecasts and responding to questions. Those workflows are particularly well-suited for AI today. When AI helps translate rough notes, emails or raw inputs into clear, structured documentation, information moves more efficiently across the project. The goal isn’t automation for its own sake, it’s reducing low-value effort so professionals can focus more time on decision-making and delivery.
Eric Pearson: The workflows that benefit most today are the ones where information is captured quickly but traditionally takes time to formalize – like meeting notes, site observations and field reports. In practice, that can mean dictating notes during a walkthrough, taking photos at the job site or logging observations on the go and letting AI turn those raw inputs into clear, structured documentation. By automating that translation step, teams spend less time writing reports after the fact and more time focused on what’s happening in the field.
Kristin McElroy: Recurring reporting workflows are a strong fit. Monthly or milestone reports often rely on the same inputs — emails, meeting minutes and project systems. When AI helps assemble those inputs into an initial draft, teams can focus their time on accuracy, context and insight instead of starting from a blank page.
Why is it important that AI supports – rather than replaces – human expertise in construction?
Jarvis Alridge: There is no replacement for professional expertise and judgment. Experience, judgment and accountability are central to successful project/program delivery. AI works best as an enabler, helping a professional process information faster, stay organized and communicate clearly. AI outputs should always be treated as draft material and reviewed by professionals who understand the project context. When AI supports, rather than replaces human expertise, it strengthens delivery while keeping responsibility and decision-making firmly with the trained professionals.
Eric Pearson: AI should be treated as decision support, not a decision‑maker. Outputs need to be reviewed and contextualized by professionals who understand the project. When used responsibly, AI strengthens delivery by improving clarity and coordination, while trust and accountability remain with the project team.
As AI becomes more embedded in delivery platforms, what should leaders keep top of mind?
Jarvis Alridge: Leaders should prioritize practical adoption over hype. The most effective use cases today are the ones that support informed decision making without disrupting how projects are delivered. When the use of AI is thoughtfully integrated into existing workflows and treated as decision support, it becomes a quiet but powerful enabler of better project outcomes.
Eric Pearson: Responsible adoption is just as important as capability. Leaders should focus on where AI can remove friction today while preparing teams for what’s coming next – including tools that can work across systems. Treating AI-generated content as draft material, reviewing outputs and following organizational AI policies enhances delivery without introducing new risks.
Kristin McElroy: Strong data governance is critical, especially as teams navigate understandable concerns about using AI on active programs. Leaders should establish clear standards for approved tools, data access and usage and data validation ownership, while reinforcing that AI-generated outputs are draft insights requiring professional review. This governance-first approach builds trust, ensures responsible use of data and maintains accountability across the program.
Mark McElroy: Leaders should also keep integration and interoperability front and center. AI is most valuable when it connects across systems—PMIS, scheduling, financials and document management—so breaking down data silos is key to getting meaningful, actionable insights.
Equally important is change management. Even the best AI tools won’t deliver value if teams don’t trust or adopt them, so leaders need to invest in training, set clear expectations and demonstrate practical use cases that show immediate benefits.
Finally, leaders should stay focused on outcomes over hype. AI should be tied directly to improving project performance—better decisions, reduced risk, increased efficiency—not just implemented for the sake of innovation.
