ConstructionGPT: AI as a Force Multiplier

There was a recent account of a mother using ChatGPT to solve a baffling medical condition that had befallen her son. She had consulted numerous physicians and received a litany of possible ailments. Finally, in a last-ditch attempt, the mother entered all of her son’s symptoms, medications, reports, etc. into ChatGPT, and the tool returned the correct diagnosis. Before all the readers decide to become armchair physicians, this was probably a unique confluence of leveraging technology and hitting the Powerball lottery. We have all gone online to investigate a bum knee or random back pain, only to leave thinking we have the Ebola virus. This is also not an indictment on the medical community since the boy’s condition was apparently extremely rare and, more importantly, the individual doctors did not necessarily have the benefit of each of the other specialists as well as the entirety of the medical community. However, the real lesson for everyone to latch onto is the potential power of artificial intelligence (AI). 

AI’s Social Media Moment

There are certainly plenty of arguments for and against AI in all aspects of society. This is not a commentary on ethics or how machines may become sentient at some point like a frightening science fiction trope. However, there is an interesting parallel for all businesses to consider —social media. 

Over the last two decades, the world has seen rise of social media such as Myspace, Facebook, Instagram, LinkedIn and X (formerly known as Twitter). As  with every new advancement, there were the early adopters who flocked to these tools and published their life chronicles with incredible fervor. Conversely, there are still businesses — since this is the lens we are using to define adoption — that fail to update their website with regularity and bristle at the thought of using social media. While this technology tool set has continued to evolve, the majority are finding a sweet spot on leverage. For many, there was the classic question of “How do we monetize social media for our firm?” Put another way: “Will we get a project or client because we are posting on LinkedIn or Facebook?” This has evolved to a more holistic approach to brand management, employee outreach, future employee connections and possibly business development/sales.

Fast-forward to the age of AI. There isn’t a day that goes by without an article, white paper, interview or advertisement extolling the virtues of AI. There is no shortage of early adopters who see this as a force multiplier, while others defiantly proclaim that Skynet will never run their businesses (thank you, James Cameron). The right answer is considering this an evolution, not a revolution. 

Leverage, Not Replacement

For all the great medical stories like the one previously mentioned, there are plenty of AI mistakes and gaffes. Just as the machine is learning, humans are realizing that human intuition and emotion are not easily replicated by a microchip. For instance, the medical diagnosis that ChatGPT returned still needed to be evaluated by a real physician with real medicine. Put another way, artificial intelligence returned a potential solution for the boy, but doctors had to apply their knowledge to enact the solution. Even with the most evolved estimating software, a smart estimator should be analyzing and strategizing appropriately. The operative word is leverage. AI has the ability to examine an infinite number of solutions quicker and provide a potential answer. Consider the following applications:

• Change order root causes — When working with specific designers, are there specific change order root causes that are more frequently attributed to errors, omissions, etc.? Are there items that seem to be routinely missed in specifications when working with specific teams?
• Work orders/client management — Are there reports that are “client facing” that seem to lack a certain level of professionalism? Put another way, are field leaders challenged to generate a meaningful and powerful service ticket that “sings”?
• Proposal management — Are proposals not resonating with specific clients? Would a fresh approach to developing a real project value proposition be a game changer?
• Cost analysis — Is there margin erosion across projects or business units, but a lack of clarity on where the loss is occurring? Put another way, other than the one-off “we left the entire switchgear package out of our bid,” what is the source of the million paper cuts that plague a firm’s ability to make money?
• Culture — Does the firm routinely survey its employees but fail to make sense of the data output with any level of accuracy? For instance, is one of the weaknesses “bad communication” but no real explanation of what this really means to the business?
• Job/role clarity — Does the organization have a real handle on what people should be doing in their roles based on the strategic direction of the business? For instance, does the organization have the standard project manager job description, but often hears things like: “But I do so much more, because we are different at Brand X Construction … ” ?

It is important to note that ChatGPT does not have all the answers to every question like a magic eight ball or mystical mirror. ChatGPT just happens to be the inexpensive alternative to a very crowded marketplace of construction-specific tools. Estimating software, drone technology and design tools are quickly adapting applications that provide the next evolution in artificial intelligence. The answer for businesses does not mean a wholesale subordination of the management of the firm to a mainframe, but careful exploration in how these tools can provide leverage, support the professionals in the firm with critical analysis and ultimately drive better decision-making. While we shouldn’t expect AI to sit in on tedious owner meetings, firms can leverage the tool to do the tedious work to allow our professionals to be greater contributors to project delivery.