Shenzhen-based Legend Robot and DaFang AI both build autonomous wall finishing robots, and both are targeting Gulf construction markets. The global construction robot market is around $1.3 billion in 2026 — small, but growing into a Saudi shortage of skilled trades measured in the hundreds of thousands.
Construction robotics has spent a decade being demonstrated and is now, in a narrow set of applications, being sold. The machines showing up in the Gulf in 2026 are not humanoids or general-purpose builders. They are single-purpose devices that do one repetitive task on an interior wall, and their commercial argument is entirely about square metres per shift.
Two Shenzhen companies illustrate the category. Legend Robot builds putty and latex paint spraying robots in 3.3-metre and 6.2-metre working heights, operated from a tablet interface that the company says requires no programming and can be run by an ordinary worker after basic training. It reports cumulative construction coverage above 10 million square metres and deployments across East Asia, West Asia, Western Europe and North America. DaFang AI builds autonomous building surface finishing robots combining machine vision with low-speed self-driving to replace manual surface treatment, operating from a research and production facility of about 4,500 square metres in Dongguan with offices in Singapore, Hong Kong and Shenzhen, and reports 128 patent filings with 81 granted.
Both are selling into the same gap. Interior finishing — putty application, sanding, priming and painting — is high-volume, repetitive, physically demanding, generates dust and solvent exposure, and produces a quality outcome that varies with the individual doing it. It is close to the ideal candidate for automation on a construction site, because the work surface is flat, vertical, indoors and repeated thousands of times in a residential tower or a hotel.
The global market is still small. Estimates of construction robots put the sector at roughly $1.12 billion in 2025 and around $1.29 billion in 2026, growing at mid-teens rates through the early 2030s. What is notable is the driver behind that growth. In every major market the case being made is labour availability rather than labour cost: unfilled construction positions in the hundreds of thousands, an ageing trade workforce and an inability to recruit at the rate project pipelines require. Suppliers have responded with narrow machines rather than broad ones, and late 2025 saw a wave of finishing-specific product launches covering high wall painting, sanding and drywall work.
Saudi Arabia's version of that gap is unusually large. The Kingdom's construction workforce exceeded 3.4 million at the end of 2025 and is forecast to keep growing to the end of the decade as giga-project delivery, the 2034 World Cup stadium programme and the Riyadh infrastructure build peak together. Against that, the sector carries a shortage of skilled workers running into the hundreds of thousands, with specialised roles staying open for three to six months on average. Finishing trades sit squarely in the shortage, and finishing is also the phase where programme slippage is most visible because it is the last thing before handover.
Whether the machines sell here will be decided by the same commercial questions that govern any capital equipment purchase in this market, and none of them concern robotics. Who owns the machine — the contractor, a specialist finishing subcontractor, or a rental company? Who maintains it, and how quickly can a technician reach a site in the Eastern Province? Where are spare parts held? What happens when the finish fails inspection: is that the contractor's defect or the supplier's? And what is the utilisation rate that makes the payback work, given that a robot idle between fit-out packages earns nothing.
Rental is the likeliest answer, as it was for most construction plant categories before it. A finishing subcontractor working across several towers can keep a machine busy; a main contractor on one project cannot. That model also solves the maintenance question, because a rental fleet operator has the incentive and the volume to hold parts and train technicians.
The broader Saudi context makes the timing reasonable. The Kingdom has been building an industrial robotics base, including a PIF-backed manufacturing venture in Riyadh producing industrial robots for export. Construction has lagged the factory because sites are unstructured. What has changed is that a small, well-defined set of site tasks has become structured enough to automate — and those are exactly the tasks arriving in Riyadh this week.