A tech and innovation executive says that the human element must be considered in this new AI landscape.
Kat Nelson Galke, chief of AI innovation & applied technologies at Trilon Group—a national AEC platform that operates across 14 companies and approximately 5,800 employees—said that while AI introduces change across AEC, Trilon is focused on turning that moment into an opportunity for enablement, growth, and more effective ways of working.
“I realized early on in my career that most of what we're doing in technology is changing a way of working,” she said. “And for somebody whose identity might be wrapped into the work they do—thinking, for example, ‘This bridge I'm building is going to be an artistic representation for a city’—that shift can be hard.
“It becomes very human and mission-driven,” Nelson Galke continued. “And now we're sitting here stating that we're going to bring agentic technology to something that could feel very much like a craft. You're adding another layer between people and the superpower they went to school for and built a career on.”
The reality, however, is that AI now sits center stage, Nelson Galke said, and will only become more prolific. She will expand on this in a keynote, The Translator Effect: Turn AI Anxiety into Advantage for AEC Firms, at PSMJ Resources’ AEC INNOVATE, June 16–18 in Las Vegas.
For Trilon, AI is also a direct response to the industry’s talent constraints, with backlog growth outpacing headcount—often at nearly twice the rate. AI represents an opportunity to unlock capacity, not replace it, she said. The goal is to create a more sustainable and rewarding work environment by reducing repetitive effort and enabling teams to focus on higher-value work. This is not about doing less with fewer people but enabling teams to do more of the work that matters most.
The session will introduce “The Translator Effect,” a leadership framework she developed for bridging business vision and technical reality. The mantra: AI will augment expertise, not erase it. Sometimes it will sit adjacent to the work, sometimes embedded within it; the use case decides which.
“It's not the great and powerful Wizard of Oz. It's literally something we're all going to be augmented by. But you can see there is trepidation and anxiety,” Nelson Galke said. AI is applied technology, she emphasizes—not a destination, but a layer that shows up inside the tools people already use. The shift, she said, is comparable to how the internet moved the industry from pulling regulation books off a shelf to typing them into a search bar: a change in how the work gets done, not what the work is.
The key will be turning that anxiety into a positive force, starting with a clear understanding of the process. “The human in the loop is imperative, especially in the AEC industry,” she said. “The question is which everyday tasks are the real pain points and where can we build in efficiencies. Can an autonomous agent handle the crosswalk between plan sets and plan summaries? Where are the operational efficiencies for a project manager?”
Not everyone will love this transition or the new roles it asks them to take on, Nelson Galke said. “They want to go and design and learn more from their community about what they can facilitate for a solution. For me, it's about bringing technology forward in a way that helps people see its usefulness on the journey—how it lets them facilitate solutions faster, better, and more efficiently. That's what compresses the anxiety.
“At Trilon, we built out an LLM that lets people bring forward ideas and surface signals of where the pain points are in the business, so we can focus on what matters,” she said. “Then our solution analysts run the discovery deep dives; they take what surfaces and ask, ‘Can this value be turned into a technology efficiency opportunity?’ And if so, is it one we build once for everyone, or scope to a specific partner firm or discipline?”
The practical version, she said, is reframing everyday work. “Instead of taking two Excel spreadsheets and manually redoing them into a third tab, what if you moved that into ChatGPT or Claude and let it do the heavy lift of crunching the numbers? Now you stay human in the loop—and your job is quality control.”
She credits Trilon CEO Mick Renshaw and CIO David Felker with directly engaging presidents across partner firms to reinforce that AI is not a one-time initiative, but a fundamental shift in how the business operates. Together, they are driving a focus on embedding these capabilities into everyday workflows, with the expectation that this impact will continue to accelerate over the next six months and beyond.
Here Are Five Action Items from Nelson Galke
Change leadership styles. Traditional leadership has to evolve to foster the new ideas that college grads are already bringing in, she said. “New technology solutions and the expectation of autonomy can feel threatening—a shift away from the operating model they built their authority on. The decision gates of the past are going to look very different in an autonomous world—and we’re going to have to find trust in those gates while keeping humans in the loop.”
Provide training. Nelson Galke built and launched an internal AI training curriculum, and in just five months Trilon trained approximately one-quarter of its workforce. She also built out a train-the-trainer program, so those trained can remain embedded in their partner firms and continue to facilitate and pass the training on. “Whether you have four offices or 14 partner firms, you need champions inside each one who can keep fostering growth and new ways of working,” she said.
Understand where the waste gap and friction are. She looks at the landscape from a business architecture viewpoint first. “What's the capability? What are the functions and tasks a person in this role is meant to do? That's the remit. Once I understand those tasks, I can see the processes and where the drag is—and solving one friction point can cut the effort in half. From there, you take that bottleneck and move it into technology enablement—that’s the step that turns the diagnosis into a solution.”
Choose your tools wisely. The tools you use need to be more than “the new cool thing to do. Is it really going to support an efficiency play?” Nelson Galke asks. “That's where many organizations get really frustrated or confused. Because you may have vendor X saying you need this, and vendor Y saying you need that, but it's not providing me efficiency. It's not going to change my backlog or automate a design element. This is where you want to de-complicate it and not buy into the AI hype.”
Make sure folks are cared for. Nelson Galke keeps coming back to how we must keep the human element top of mind. “It has to be. When you're thinking about the trust that these folks are going to need to lend to the technology. This is a new mental model and potentially an identity shift that will likely bump heads with passion and perceived superpowers. We need to care for the humans who are on the other side of this. AEC is one of those older traditional industries, and we need to figure out how we are going to journey and walk through this transition.”
This is content from the PSMJ Journal, exclusive to PSMJ PRO Members. PSMJ PRO is the fastest-growing network of AEC firm leaders. Not a PRO Member? Register here