When OpenSpace announced in early June that it had surpassed 1,000 data center projects globally, the number landed with a particular weight. Not because 1,000 is a round number — though it is — but because 500 of those projects were added in the past year alone.
The acceleration tells a story about two intersecting forces. One is the AI infrastructure buildout: the unprecedented wave of data center construction being driven by cloud compute demand, LLM training clusters, and hyperscaler expansion. The other is the maturation of construction intelligence as a category — the point at which tools like 360° photo documentation stop being an operational nicety and start being a project standard.
OpenSpace has arrived at the intersection of both.
What OpenSpace Does
OpenSpace makes a construction documentation platform built around a 360° camera that attaches to a hard hat. Workers walk the job site as they normally would; the system automatically captures continuous 360° imagery and stitches it into a navigable, timestamped record of site conditions. No manual photography, no dedicated documentation crew, no selective coverage of the areas someone remembered to photograph.
The resulting record can be compared to BIM models, overlaid with floor plans, and reviewed remotely by anyone on the project team. A project executive in New York can walk a site in Austin without leaving their desk. A subcontractor trying to understand existing conditions before mobilisation can review last week’s site state from the office.
That is useful on any project. On a data center project, it is close to essential.
Why Data Centers Are Different
Data center construction has characteristics that stress-test conventional construction documentation. The density of systems in a raised-floor data center — power distribution, cooling infrastructure, fire suppression, structured cabling — is extreme. The interfaces between trades are everywhere, and errors at those interfaces are expensive. Access is restricted: not every member of the project team has clearance to be in certain areas of an active data facility, which means documentation has to substitute for physical presence.
The schedule pressure is also unusual. A hyperscaler building data center capacity to support a product launch or a contracted compute commitment does not have the flexibility for the kind of delay that documentation failures can cause. When a conflict emerges because nobody documented which conduit runs were installed before the concrete pour, the downstream effects cascade fast.
OpenSpace’s platform addresses these pressures directly: continuous documentation removes the gaps that disputes fall into, remote access extends the team’s visibility without requiring physical presence, and the timestamped record becomes a forensic resource when something goes wrong.
The company has been specific about the data center use case for some time, but the milestone announcement is the first time it has quantified the scale of its presence in that vertical. One thousand projects globally, across the hyperscalers, their general contractors, and their subcontractors, is a substantial installed base. More importantly, 500 of those projects in the past year reflects a rate of adoption that is still accelerating.
The Dataset Question
There is a dimension of the OpenSpace data center milestone that the company has been careful not to overstate but that investors and competitors are almost certainly thinking about: 1,000 documented data centers represents an extraordinary training dataset.
Every OpenSpace project generates structured, timestamped, georeferenced 360° imagery of a construction site as it progresses from ground condition to practical completion. Over 1,000 data center projects, that represents an enormous volume of visual documentation of how data centers are actually built — the sequence of trades, the typical conflict points, the patterns of rework, the ways in which schedule delays manifest in site conditions.
That data is the raw material for a generation of predictive and analytical AI tools that the industry does not yet have. A system trained on sufficient site imagery from data center projects could, in principle, learn to predict where coordination failures are likely to occur, identify installation sequences that correlate with rework, or flag site conditions that have historically preceded schedule delay.
Whether OpenSpace will build those tools internally, partner to build them, or treat the dataset primarily as a retention and expansion asset is an open question. But the fact that the company can now describe its data center coverage in terms of thousands of projects changes what is possible.
The Broader Contech Moment
The OpenSpace milestone is a useful lens through which to view the broader construction technology landscape in mid-2026. A year ago, the conversation around AI in construction was heavily weighted toward what AI might eventually do. The discussion was prospective and theoretical: agents that could read contracts, robots that could lay bricks, systems that could predict schedule risk from project data.
The 1,000-project figure is a reminder that the most impactful construction AI deployments right now are not the most speculative ones. They are the tools that reliably solve a real operational problem — in OpenSpace’s case, the problem of knowing what is actually happening on a job site at any given moment — and that have been deployed at sufficient scale to generate the data required for the next generation of capability.
The data center construction boom is not slowing. Hyperscaler capex on infrastructure has continued to grow through 2025 and into 2026, driven by compute demand that remains structurally underserved. Every new data center project is a potential OpenSpace deployment. The question the company is now positioned to answer is whether it can use the embedded position it has built in the sector to expand from documentation into the analytical layer — and whether it can do so before the platform players build documentation features into their own suites.
