For the past three years, the conversation about artificial intelligence in construction has been dominated by images of robot bricklayers, drones over site and generative designs that look like coral reefs.
Spinifex is an opinion column. If you would like to contribute, contact us to ask for a detailed brief.
They make good conference slides, but they are not, for the most part, where the change is happening. The less glamorous truth is that the real shift is taking place in the paperwork, and it matters more than the hype suggests,
Consider what construction actually runs on. A major project can generate tens of thousands of documents, including contracts, cost plans, programmes, change requests, minutes, specifications and reports. Much of the professional effort in our industry goes into reading, reconciling and rewriting that material. This is precisely the work that large language models do well, and it is where adoption is quietly accelerating.
The evidence reflects this, with an interesting twist. Research by the Association for Project Management (APM) and Censuswide, published at the end of March 2026, found that 28 per cent of UK construction project professionals now describe AI as fully embedded in their workflows, the highest among the sectors reported. Yet only months earlier, APM’s survey of business leaders ranked construction last among the ten sectors for productivity gains, also at 28 per cent. How can both be true?
The answer tells us where the industry is heading. While construction adopted the tools quickly, it largely bolted them on to existing ways of working. An assistant that summarises a contract in thirty seconds saves time for the individual, but it does not change the process, the fee model or the outcome for the client. The productivity dividend only arrives when firms redesign the workflow around the technology rather than the other way round.
In my view, there are three trends that will define the next phase.
The first is the move from chat to agents. The early wave of AI was conversational. A person asks, the machine answers. The next wave consists of systems that carry out multi-step tasks, including drafting a monthly cost report from live data, checking a contractor’s program for logic errors, assembling a tender comparison) with the professional reviewing rather than producing. This changes the shape of a consultant’s day and, in time, the shape of a consultancy.
The second is that rather than models, data becomes the constraint. The models are now broadly capable and increasingly interchangeable. What separates the firms from gaining value from those is not the quality, structure and accessibility of their own information, but the historic cost data, lessons learned, project records. Network Rail’s data first approach, documented in an APM case study, makes the point well. Organisations that neglected their data for decades are discovering that it is their most valuable asset, and that it is in poor shape.
The third and perhaps the most important is sustainability. AI has real potential, including estimating embodied carbon at concept stage, finding out when decisions are cheapest to change, optimising building performance in use, and making sense of the reporting burden that ESG now places on every project. However, AI carries its own footprint in energy and water. An industry that asks clients to account for whole life carbon should apply the same discipline to its own tools, using AI where it earns its keep rather than everywhere by default.
There is also a larger prize. A hope is that as these systems move towards implementing artificial general intelligence, which can help us solve some of the most significant sustainability problems facing society. These include low-carbon cement and steel, grid-scale energy storage, and the modelling of climate risk across whole cities. If that comes to pass, the energy AI consumes today may prove a sound investment. It would be unwise, however, to count on it.
While none of this removes the need for professional judgement, it raises the premium on it. When a machine can produce a plausible cost plan in minutes, the value lies in knowing whether it is right, what it has missed and what the client should do about it. APM’s latest research found that professionals rank ethical decision-making and professional judgement at the top among critical future skills, alongside data literacy. I agree. The risk is not that AI replaces the quantity surveyor, the engineer or the project manager. It is that we are training a generation to accept outputs they are no longer equipped to question.
So where does that leave us? The robots may come in time, but the transformation already under way is quieter and closer to the desk. The firms that thrive will be those that treat AI as a reason to rethink how they work, invest in their data and their people, and hold the technology to the same standard of accountability they would expect of any other member of the project team.
That is less exciting than a robot bricklayer. It is also far more likely to change the industry.
