In a profession requiring precision and accountability, the adoption of AI only works if it is trustworthy, transparent and integrated into familiar tools and workflows.
Across engineering and construction, an appreciation of the potential benefits of AI is developing fast, but so is hesitation.
As engineers are tasked with delivering increasingly complex projects under tighter timelines and within a more restrictive regulatory regime, they recognise the enormous benefits AI could bring. However, unless the tools it powers can be interrogated, explained and easily adopted, traction is unlikely.
“In the world of engineering, there is a lot of noise, regulation, sustainability, building codes, a lot of work that is piled on top of the raw engineering that engineers did not necessarily get into the profession to do,” said Jason Howden, Chief Innovation Officer at Revizto.
It is in these areas that AI has the potential to remove much of the burden, Howden said. But it will only work if it meets the same standards as other engineering tools. Black box systems – software and tools that produce outputs without transparent reasoning and that can’t be tested or validated – offer limited value to engineers.
Howden said AI should never displace that responsibility. “It’s not about taking the engineers out of the room,” he says. “Success is about allowing engineers to make really good decisions in a time-crunched industry, with confidence.”
That type of confidence can only come from transparency, and from how effectively the technology that delivers AI fits into existing engineering workflows.
Removing friction from workflows
Some AI offerings promise transformation while introducing extra complexity and learning requirements. But as engineers are already juggling a wide array of platforms and tools, formats and stakeholders, what they require is better integration.
“For us, AI provides a natural way to talk to the information already in Revizto,” Howden said. “That means you don’t need a degree in how to use new software, you just now have a much more simple, natural and thorough way to access information you’re already using.”
“Think of the digital information currently captured in construction. Integrating AI into the systems engineers are already using will give those engineers superpowers to lean into their back catalogue of intelligence and IP.”
Practical applications are already appearing in the field, demonstrating the shift from manual to AI-driven engineering.
For example, Howden said, instead of manually interrogating models, drawings or spreadsheets, engineers can query the project data directly. A construction manager, for example, can ask how many bolts are required for a project and receive an immediate, verifiable answer drawn from the project’s own data sources.
Similarly, a designer could ask for a report on embodied carbon across a project, then use that insight to inform design decisions in real time.
These use cases are not science fiction and nor are they abstract. They reflect common pain points in digital engineering including coordination, visibility and decision-making under pressure.
Another example of AI in action is the dramatic enhancement of the clash detection process. On a project, simple or complex, engineers can run automated checks across models to identify conflicts. Layering AI on top of this, Howden said, allows systems to recognise not just clashes within the current project, but also to analyse patterns from previous projects to suggest solutions or alternatives.
“We know what worked in the past and what didn’t, because we’ve got this track record at our fingertips,” he said. “Being able to surface relevant information faster and reduce manual effort enables more informed decisions and even better engineering judgement.”
Data sovereignty and control
Working with an organisation’s or project’s own data, that which already exists and is regularly updated and added to as things move forward, removes a major barrier to AI adoption.
When it is integrated into current engineering processes and technologies, AI operates within the organisation’s usual boundaries. That means the authority and reliability of the data is certain, and that confidentiality and commercial value is protected.
AI introduced in such a way, therefore, minimises new risks. Instead it brings better, more thorough and efficient interrogation of the organisation’s data.
It also enables engineers to continue to use their preferred models and processes.
“Locking data and workflows into a single proprietary system risks long-term dependency and reduces adaptability,” Howden said.
Expertise is boosted, not replaced
The most valuable role for AI in engineering is not automation, but instead augmentation, Howden said.
It reduces the administrative burden surrounding engineering and in doing so enables engineers to focus on design, problem-solving and innovation.
This shift has broader implications beyond individual project requirements. More efficient workflows translate into better outcomes for clients and end users through improved design, reduced rework and better use of resources and materials.
“We’re all creative problem solvers,” Howden says. “If we remove the noise, it allows us to be more creative around solving these problems, and that’s good for everybody.”
Explore Revizto’s AI Manifesto.
