This article has been provided by Aly MacGregor, property executive at WSP

Have you heard the one about an AI that couldn’t create an image of a bike with square wheels, since all its training data had bikes with round wheels? You could take that as a humorous illustration of AI’s limitations – or a cautionary tale of the limitations of human engineers. 

Most engineers I know take a degree of pleasure in demonstrating how clever they are, in this case outwitting AI. Yet, I question whether that’s the best demonstration of intelligence. After all, breaking a model isn’t necessarily that hard, if that’s your goal. We would be far smarter and better served to figure out new ways to use AI to address more meaningful questions. 

Instead of figuring out how to break a model, engineers should be asking how to optimise design, improve efficiency, or solve complex challenges. For example, rather than ask about bikes with square wheels, we should be asking about practical challenges, such as creating bikes that use more sustainable materials to improve the weight to power ratio. By asking more creative questions, we can give AI the chance to provide a more creative and valuable response. 

Unlocking that potential requires a fundamental shift in the way engineers think about how and why. 

Causation and creativity 

From Aristotle to Newton to Othmar Ammann and Charles Whitney, engineering has been built on the application and interpretation of rules about why things work the way they do. Deeply analysing why something happens is at the heart of every engineer’s career journey since causation is a fundamental of our field. 

We are taught to know why “galloping Gertie” swayed, why the space shuttle exploded, and why power lines experience loss of load. More recently, my colleagues and I have asked why floors can’t generate electricity from foot traffic, why a seismic zone can’t have a floating bridge for cars and trains, and why skyscrapers can’t be made of wood. (Turns out, they can.) 

But what if we balanced our focus on why with a greater focus on how? That’s where AI can help free us to become even smarter. 

For example, an engineer might consider solar radiation, air pressure, precipitation, and all the other drivers of weather. Yet, most people – and clients – just want to know whether to wear a jacket or bring an umbrella. That’s why engineers need to embrace a new balance of focus, knowing when to dive deep into the why and when it’s paramount to get to the how as quickly as possible. 

Enter your AI design partner. 

Since most current AI tools don’t get hung up on causality (the why) and can mine accumulated knowledge and previously developed solutions to present a range of “hows,” AI doesn’t have an ego that needs affirmation. The tool trusts the work of others (which we do need to check) and can provide impartial solutions. AI and Large Language Models (LLMs) are (or soon will) challenge traditional approaches to design and problem-solving. Since most engineers stereotypically don’t like change, these shifts might not be easy – but this challenge could be a fun one. 

How can we shift to the how? 

First, let me be clear: I’m not recommending a loss of the curiosity, exploration, or questioning of assumptions that are at the heart of who we are as engineers. Rather, I think we can see this as an opportunity to work in new ways, bolstering our focus on the areas where we can deliver the greatest value. 

A useful analogy is the auto industry. For many manufacturers, 80% of the base vehicle is the same across models and finishes. But the differentiating 20% enables the company to market brands that speak to very different customer needs and values. AI could help us take a similar approach in engineering. 

By elevating the importance of operational insights, we can ensure that actual performance—not theoretical performance—drives decisions around “how,” making our recommendations more credible. In practice, that could mean tapping AI for repeatable design efforts. By focusing a little more on the “what,” we can realise opportunities for a greater level of design consistency across projects and allow us to capitalize on the benefits of greater productization.

We’re already seeing similar benefits from our kit-of-parts approach, which is using technology to codify knowledge and helping build a “memory” for each project. It helps address labour and skills, enabling teams to make small decisions once – and allowing them to move faster and freeing them to focus on bigger issues, such as circularity. 

If we apply this approach early in the design phase, it could help ensure our projects are more future ready, as we deliver credible virtual prototyping insights to help inform decisions today that could be impacted by potential future conditions. By tapping AI, we also have another tool for QA/QC and constructability evaluations of design solutions. 

New human intelligence 

This transformation doesn’t just require AI; it will require new kinds of human intelligence, too. The integration of AI will demand that engineers enhance our ability to navigate between deep analytical thinking, rapid problem-solving and new ways to translate stakeholder needs into technical terms. We’ll also need to know everything from how to provide effective prompts to the ability to critically evaluate AI-generated recommendations. 

If we get this right, AI could be the impartial design partner engineers need, helping us make more informed, efficient decisions while preserving the core spirit of engineering: understanding, innovation, collaboration and problem-solving. 

The future of engineering lies not in resisting technological change, but in embracing and shaping technical change – and how!