The AI Opportunity Gap: From Data to Insights to Decisions
Most AI tools in AEC aren’t actually helping you make decisions. They’re just organizing data.
The AEC industry talks a big game about AI, but too many tools are stuck in the past — focused on tracking history rather than shaping the future. If we want AI to deliver on its promise, we need to stop asking architects and engineers to become data scientists first.
In the world of AEC, innovation often struggles against ingrained habits and outdated processes. While the industry eagerly discusses the transformative potential of AI and ML, most current solutions remain tethered to the past, primarily focused on counting, accounting, and reporting. To truly redefine productivity, we must shift our focus from merely auditing history to actively monitoring the present and proposing actionable solutions that will take us where we will be in five to 10 years.
At a recent BuiltWorlds event, discussions among venture capitalists, software startups, and AEC firms highlighted and underlying industry-wide apprehensions: uncertainty about AI’s role in reshaping workflows. The prevailing conversation still revolves around capturing and structuring data rather than leveraging AI to drive real-time, in-the-moment decisions. If AI is to fulfill its promise of transforming productivity, the industry must stop expecting firms to become digital experts just to make AI work for them.
The Stages of AI Maturity in AEC
Currently, most AI tools focus on structuring and sorting data rather than generating actionable intelligence. If we sort AI-driven solutions four stages of maturity, we get:
1. What Happened?
Many existing tools are geared toward creating audit trails— submittal logs, RFI logs, and change order logs. These are helpful for understanding past events but have currently done more to mitigate legal risk than influence ongoing processes.
2. Why Did It Happen?
The next logical step is understanding causality. AI tools are beginning to analyze patterns in project data, helping firms discern the root causes of delays, cost overruns, and inefficiencies. In the short term, AIs will start inferring potential answers for professionals to consider rather than explicitly executing actions.
3. What Is Happening Right Now?
This is the crucial missing link. Few solutions effectively track real-time project status in a way that enables immediate decision-making. Without this, AI tools cannot reach their full potential in impacting AEC outcomes.
4. What Should I Do?
True AI-driven transformation lies in prescriptive analytic tools—providing actionable insights before problems arise, not after they’ve already disrupted a project. We may be some way from this being real, but it is an inevitable future solution set.
As inferred above, I believe too many software companies are placing the burden of data strategy on AEC firms, expecting them to become digital experts and data strategists before AI can begin to help them. This approach may be flawed as it will be incredibly difficult to have a consistent base line of data strategy/format across a multitude of firms, such that a software supplier can easily plug in their solution. Currently, too many AEC software companies ask that firms must first create a comprehensive data strategy before AI can deliver value.
The result? Many AI applications in AEC remain immature, focusing more on accumulating and organizing data rather than generating actionable intelligence.
Instead of advancing the conversation about what AI and machine learning should enable — real-time decision-making and predictive solutions — many discussions remain stuck on how to move more data into software.
By comparison, most of the successful Silicon Valley companies never required users to organize their data before offering solutions. Facebook didn’t ask users to curate their photo libraries before posting, and Google didn’t demand website owners structure their data before indexing the web. AEC AI solution providers might spend more time providing their intelligent, real-time guidance by ensuring their solutions also do more of the heavy lifting on parsing data.
The industry needs a shift. AI solutions should reduce complexity for AEC professionals, not add to it. The real opportunity lies in developing tools that provide insights without requiring firms to become data scientists. Moving forward, we need a more sophisticated dialogue — one that prioritizes AI’s role in shaping the industry’s future rather than simply documenting its past.
The Opportunity Gap: Real-Time Decision-Making
Despite AI’s promise, the industry largely lacks solutions that can inform decisions while projects are still in motion. For example, a tool that predicts a concrete truck’s arrival time in relation to impeding rain could prevent costly rework. Companies like Trunk Tools are beginning to address this gap by enabling better interaction with existing documentation, but comprehensive AI-driven workflow optimization remains elusive.
Another significant discussion at BuiltWorlds centered on the challenge of feeding real-world project learnings back into the design phase. The concept of a “digital twin” is often mentioned but remains poorly defined and inconsistently executed. AEC professionals must ask themselves:
Are there enough incentives for firms to capture valuable data at each stage in a usable format?
Automation’s Role in Elevating Human Creativity
One area where AI can make an immediate impact is eliminating repetitive tasks that bog down young designers. In many firms, early-career architects spend years performing tedious, manual tasks as a badge of honor, of paying your dues. AI-driven tools like Skema can help automate these processes, allowing designers at all stages of their careers to focus on creative problem-solving rather than routine drafting.
Yet, a cautionary note emerged during the conference: AI’s reliance on large datasets may push the industry toward uniformity. If AI only generates designs based on what has been done before, it risks producing the lowest-common-denominator solutions. Instead, AI should be leveraged to explore unique, differentiated approaches — offering architects a competitive edge rather than reinforcing mediocrity.
Moving Forward: AI as a Partner, Not Just a Reporter
For AI to truly revolutionize the AEC industry, it must move beyond static documentation and toward dynamic, real-time decision-making. AI tools that simply aggregate existing data will drive firms toward the safest, most common solutions rather than fostering innovation. AEC needs AI that delivers differentiated, high-value insights — helping firms make better decisions, not just organize past ones.
The future belongs to tools and systems that can:
Provide live insights to inform critical decisions mid-project.
Enable seamless round-tripping of lessons learned into future designs.
Empower architects and engineers by automating mundane tasks, allowing them to focus on innovation.
The AEC industry’s journey with AI is still in its early stages. Much like tadpoles struggling onto the mudflats, we are navigating uncharted territory. The firms that break free from outdated habits and embrace AI’s potential to be a true partner in decision making will lead the charge into a more efficient, innovative future.