Rework costs the UK construction industry around 11% of total project costs. Some studies put it higher. Between 18% and 25% of project costs in certain sectors, according to research cited by PlanRadar. In every case, the root cause is the same: defects that were not caught early enough, or not caught at all before the next trade moved in on top of them.
The traditional answer was the clerk of works. An experienced professional walking the site daily, checking workmanship against specification, flagging issues before they became problems. That role became a cost casualty in many projects during the 2000s and 2010s. AI is now offering something that might do a version of the job continuously, at scale, without the day rate.
What the platforms do
OpenSpace allows site teams to capture 360-degree walkthroughs using cameras worn on a hard hat. The footage is automatically mapped to project plans and timestamped. AI tools then identify changes between walkthroughs, flag issues against design intent and route them to the relevant trade for resolution. OpenSpace has now captured 64 billion square feet of construction across its global user base. Its OpenSpace Field product, launched in early 2026, brings this directly to smartphones, with AI voice notes turning spoken observations into structured task data.
Buildots takes a similar approach but focuses on progress tracking and delay prediction. 360-degree cameras capture site conditions. AI compares what is on site against what the programme says should be there, generating deviation alerts and production rate calculations. The system can predict construction delays before they become visible to the human eye.
The Cost of Getting It Wrong
11%: Rework as a share of total project costs across UK and EU (PlanRadar, 2026)
18-25%: Rework share in some UK studies
52%: Reduction in rectifications reported by PlanRadar customers
30 mins: Inspection time reduced from 3-4 hours by OpenSpace Field (T&H Investments)
Real examples in UK construction
UK contractors are piloting these tools on major projects. Buildots has been used on hospital builds and commercial schemes where programme adherence and quality compliance carry significant contractual consequences. OpenSpace has been deployed by main contractors and subcontractors alike, with the documentation trail proving valuable for dispute resolution as much as for quality management.
The PlanRadar 2026 rework study, which surveyed over 2,500 customers across the UK and Europe, found that poor communication between teams and stakeholders was the top cause of rework in the UK, followed by quality control failures. Both are precisely what AI documentation tools address, by creating a shared visual record that all parties can access in real time.
"The field is the most underutilised source of project intelligence in construction. Closing the gap between seeing an issue and acting on it is where real time savings are made." OpenSpace, 2026
Human oversight is not optional
These tools are powerful. But treating them as a replacement for human judgment is a mistake. AI visual inspection can identify a crack in a wall. It cannot tell you whether that crack reflects normal thermal movement or a structural problem that requires immediate remediation. That still requires a professional with expertise, standing in front of it.
The Building Safety Act has made this particularly important. Fire safety defects in residential buildings over 11 metres involve an estimated remediation cost of £3 billion. Many of those defects went unchecked at construction stage. AI inspection tools that are properly integrated into quality workflows, with clear human sign-off requirements at each stage, could help prevent the next generation of building safety crises.
The responsible deployment question is about role definition. What does the AI flag? Who reviews it? Who signs off that an issue has been resolved? Without clear answers, you have a documentation tool, not a quality assurance process.
Used well, AI inspection closes the information gap between what was built and what was specified. Used carelessly, it just creates more data for the dispute that comes later.
Building AI Competence in the Built Environment
The Responsible with AI programme gives built environment professionals the tools, frameworks and practical guidance to govern AI confidently. Including quality assurance governance, human oversight requirements and AI tool procurement for construction teams.
Explore the Programme → Responsible with AI
