AI adoption in built environment ‘requires mindset shift’

The adoption and use of AI in the built environment sector is growing, with around a quarter of organisations surveyed by Pinsent Masons and industry leaders already taking active steps to adapt their business models for significant future disruption. However, the feedback also shows a lack of readiness in the sector to adopt and use both general and bespoke AI solutions, which is holding it back from embracing the full suite of productivity gains that AI can bring.

Pinsent Masons along with Bentley Systems, Mott MacDonald and Turner & Townsend sought the views of a wide range of organisations across the built environment sector globally between February and June 2025. Respondents included architectural and engineering design consultants, contractors, project and cost managers, technology providers and asset owners and operators.

Of those organisations surveyed, 24% identified AI-driven change as a major issue their businesses were currently grappling with and said they have already taken active steps to adapt their business models in anticipation of significant future disruption. Although 40% of respondents anticipate that AI will impact their business models significantly in the coming years, only 20% said that AI is of current strategic importance to them or that they have plans to adopt it more widely across their organisations.

When it comes to AI use, the most common uses reported by respondents were in automating document-related processes; real-time monitoring of operational data to predict when maintenance is needed or to optimise and improve the performance and resilience of infrastructure and building assets; and optimising construction productivity. Where organisations are already investing in AI, there is already a strong focus on leveraging AI systems to optimise and improve the productivity of design and engineering processes.

When asked about their specific use of AI to date, 36% of those surveyed reported using generic AI systems such as Copilot. While only 15% of respondents said they had selectively used AI or trialled applications and were yet to integrate the use of AI into their operations, 26% said their businesses were already using and trialling AI systems with specific use cases in mind.

The use of agentic AI systems – which leverage ‘static’ AI agents and other tools and resources to build additional capabilities, like reasoning, planning and self-evaluation – appear not to be widely used in the sector yet. However, only a small proportion of respondents said they were not actively considering adopting or using any AI at all.

Ian Laing, an infrastructure and real estate expert at Pinsent Masons, said the responses should give organisations across the built environment sector pause for thought about where and how AI can be deployed most effectively. “Optimising and improving the productivity of design and engineering processes is one current area of focus,” he said. “This includes generative design, complex simulation, digital twinning and rapid analysis of solutions which can help optimise design and engineering solutions.”

"Other areas of focus include automating document processes and cost estimation and forecasting."

A sizeable proportion – two-thirds – of architectural and engineering designers surveyed are either already taking active steps to adapt or see significant future disruption to their current operational business models.

While there have already been advances in using AI-powered automation in sectors such as mining, the construction industry, particularly construction contracting, has historically been much slower to change and embrace new technologies. This is borne out in the survey data, which underscored noticeable reticence and scepticism from contractors towards AI. Only 12% of contractors are actively taking steps to adapt their current business models for future disruption. While 35% of contractors expect AI to have a significant impact on their business model in the next three years, over 50% currently anticipate that AI will have a more limited impact or won’t have any impact at all.

Graham Robinson, an infrastructure expert at Pinsent Masons, said AI has the potential to disrupt even core construction tasks sooner than many contractors might think. “While practitioners are sceptical about on-site automation in the short-term, the long-term potential remains high and contractors that ignore this potential may become casualties of the rapid advancement of AI,” he warned. “Contractors could pilot AI and robotics in safe, incremental ways now – such as autonomous drones for site surveys or AI for safety monitoring – so as not to fall behind when on-site automation accelerates.”

Of the survey’s 86 respondents, 75% were at c-suite or boards level across the built environment sector, with input primarily from the UK, North America and Australia, as well as some responses from Europe, Asia, the Middle East and Africa.

This makes the survey’s findings around governance and risk all the more compelling, the experts said. While 60% of those organisations surveyed said they had some kind of AI policy in place, 20% indicated they had implemented a general organisation-wide AI policy. Just 6% of organisations said they had specific policies in place for individual departments, setting out guidelines for AI use, governance, ethical implications, safety measures and related aspects, with plans to update policies. A further 12% said they had no department-specific or organisation-wide AI policies in place at all.

Some 37% of organisations said they currently have limited or no project controls to manage risk where AI is used for design and construction. Such gaps could expose organisations in the built environment to further unwanted risks and liabilities, said Robinson. “Without adequate governance and project controls, AI mistakes could pose safety risks or legal liabilities,” he said. “AI model validation, peer review of AI-assisted designs, and clear policies and contractual terms about AI use will help ensure that there are appropriate checks and balances.” He said regulators may also need to update guidelines to cover AI-assisted engineering.

Looking ahead, more than two thirds of respondents indicated that they expect over half of their projects to significantly leverage AI in the design and construction of infrastructure and buildings within the next three years. A total of 29% of those surveyed expect similar gains to be delivered in construction works, suggesting there is still considerable scope to expand the AI use cases in this section of the market.

Global construction spending is currently projected to hit $10 trillion in 2025, according to Oxford Economics. As the disruptive and transformational impact of AI is expected to gather pace and momentum in the next few years, Laing said the built environment sector must be poised to adopt, use and invest in bespoke AI solutions to maximise productivity gains in an increasingly unpredictable business climate. “Significant levels of investment will be needed to change a siloed and heavily fragmented industry to a more integrated, digital data-driven, and modernised alliances of different capabilities,’” he said.

Robinson said a mindset and cultural shift, coupled with significant investment in digital skills and training, will also be critical to help the sector keep pace with AI momentum. “Expecting a jump to a majority or higher levels projects where AI is significantly leveraged in three years will require a significant shift in the construction industry’s conservatism and culture and the actual pace of adoption may be slower if challenges are not addressed,” he said. “The bridge between readiness and maturity, that we found is evident, will need to be addressed. Challenges such as skills employment and the need for much greater data standardisation are major considerations.”