Picture credit: Solen Feyissa on Unsplash

So far, 41% of UK architects have used artificial intelligence (AI) on at least an occasional project. Of those architects, 43% think AI has made the design process more efficient. These numbers are expected to increase, according to new research from the Royal Institute of British Architects (RIBA).

More than half of architects (54%) are expecting their practice to be using AI within two years, and 57% think it will improve efficiency. My view is that, while I have a great deal of respect for RIBA and this is valuable research, we need to be careful when people are “self-reporting” on whether they’ve used or plan to use something.

It’s a little like people telling their doctor they consume less than 14 units of alcohol each week – some are being truthful, and others are bending the truth or being overly optimistic.

The research also shows that 69% of respondents had not yet invested in AI research and development, and only 41% expect their practice to invest in this. My assumption is that some architects have dabbled in using AI to “test it out” but most are not seriously putting AI to work, because it takes time to figure out which tools will be beneficial, and where in the process it’s best to apply them.

Digital maturity in architecture

When asked about their organisation’s digital maturity, 47% of architects felt they were about the same as other organisations in the sector. That is, they show a willingness to adopt new digital tools that are already of proven value.

This backs up my theory that until it’s easy to know how to use AI in a positive way, adoption will be slow. The innovators and early adopters in the sector will spend time and money figuring out the best ways to put AI to use, and then this information will very gradually emerge until we reach a tipping point of widespread adoption.

As RIBA’s report points out, the discussion around AI is intense. From “existential despair” at the potential for AI to cause the loss of architecture and other professions completely, through to celebration of human creativity and a feeling that “no machine could create as we create” – there is no shortage of opinions.

But there has been a shortage of facts and evidence. RIBA’s research is a useful starting point to give us a sense of what’s happening on the ground.

The pros and cons of artificial intelligence

AI has the potential to automate routine tasks and strip out some of the boring work, leaving us with more time for work that’s challenging, creative and enjoyable. It could also help to streamline the design process and reduce pressure on resources.

Of course, there are risks if we hand over work to AI. If some work is done by AI rather than humans, will fees be pushed down? Will clients expect more for less if they know that AI is picking up some of the workload? Will architecture practices downsize once AI is doing a chunk of the work? These are real fears, but it is going to be some time before we see a level of AI adoption that turns these fears into reality.

How are architecture practices using AI?

The most common use for AI is during the early stages of the design process, and the second most common usage is for project management. For early design stage visualisations, 6% of respondents always use AI, 22% often use it, 60% sometimes or rarely, and 12% never use it for this purpose.

For generative design, 21% always use AI, 31% use it sometimes, 16% rarely and 32% never. AI is also being used in the following areas:

• Parametric design,
• Model generation,
• Building performance simulation,
• Environmental impact modelling,
• Specification writing,
• Standards and regulatory compliance checking,
• Construction product and material selection and analysis.

Overall, the impact of AI on the construction industry is seen as positive by survey respondents. With over 500 RIBA members surveyed, RIBA themselves say that while the survey results give a good indication of how AI is used and seen within the profession, they’re not definitive.

The risks of AI

The risks are vast. As one survey respondent puts it, “There are no real regulations in place and the ethical risks are very significant, from intellectual property, design creativity, employment and potential risks in the built environment too (if things go wrong).”

There are some valid points here, especially when it comes to intellectual property, but it seems unlikely we’ll ever move to a scenario where no human is double-checking the work of AI. If we do reach that point, then we’ll have something to worry about.

What do you think about AI in architecture? Are you using AI and, if so, what are you using it for? I’d be interested to hear from you.

In the meantime, if you need assistance with the structural elements of an upcoming project, please do get in touch.