Dapple targets governed AI cloud gap

Dapple has secured $30 million in seed funding to expand a dedicated enterprise AI cloud platform aimed at organisations that want production-grade artificial intelligence capacity without relying on shared public cloud environments or slow private data-centre builds.

The round, backed by The Raptor Group and Ion Pacific, gives the company fresh capital to scale its Enterprise OS Cloud, a single-tenant infrastructure model designed for regulated companies, governments and AI-native businesses that need control over data location, performance and governance. The financing comes five months after Dapple’s launch, with the company saying it has already signed more than $100 million in customer contracts and has live production workloads running across two continents.

The funding highlights a sharper shift in enterprise technology spending as AI projects move from pilots into heavier workloads that require graphics processing units, high-speed networking, predictable latency and stricter compliance controls. For banks, healthcare groups, pharmaceutical companies, public-sector agencies and large industrial users, the main concern is no longer only access to AI models. It is where sensitive data is processed, who controls the infrastructure, how capacity is reserved, and whether performance can be audited.

Dapple’s platform is being positioned between the hyperscale public cloud and a fully owned private data centre. The company says its model provides dedicated, in-country GPU infrastructure operated through a unified control plane, giving customers the isolation associated with private infrastructure while reducing the time and complexity of building from scratch. Its public material says deployments can go live in three to nine months, a timeline aimed at enterprises that cannot wait through multi-year internal infrastructure programmes.

Chief executive Tricia Martínez-Saab has described the company’s thesis as a response to a gap faced by enterprises with demanding AI workloads: shared cloud environments may not meet isolation and residency needs, while private builds can be slow and capital-intensive. Co-founder and chief operating officer Salam Al-Mosawi brings experience in AI and cloud compute ventures, with the company’s leadership emphasising GPU deployment, data-centre strategy, enterprise sales and partnerships as core areas of execution.

The company’s Microsoft Marketplace listing describes Dapple as an Azure-native Private AI Cloud for regulated and mission-critical workloads. The platform is designed to extend Azure services into dedicated, single-tenant, in-region GPU infrastructure, allowing customers to use familiar tools while gaining stronger control over execution, governance and location. That positioning may help Dapple appeal to enterprises already standardised around major cloud platforms but seeking more control than standard shared capacity provides.

The timing is favourable for specialist AI infrastructure providers. Demand for accelerated computing has strained global data-centre capacity, while large technology companies and AI labs have been signing multibillion-dollar compute arrangements to secure power, chips and networking. At the same time, governments and heavily regulated sectors are placing more emphasis on data sovereignty, local processing and auditability. The result is a growing market for hybrid, private and sovereign AI infrastructure that can support model training, fine-tuning and inference without forcing sensitive workloads into generic shared systems.

Dapple is not alone in targeting this opening. Hyperscalers are expanding sovereign cloud offerings, neocloud providers are renting GPU clusters, and hardware vendors are promoting “AI factory” models that combine dedicated compute, storage and networking. Enterprises also have the option of building on-premise environments, though that route often requires specialised talent, long procurement cycles, energy planning and ongoing operational discipline. Dapple’s challenge will be to prove that a managed single-tenant model can combine the speed of cloud consumption with the assurance of dedicated infrastructure.

The company’s early contract figure will be watched closely because infrastructure start-ups often face a difficult balance between sales momentum and capital intensity. Dedicated AI environments require access to GPUs, data-centre space, power, cooling and network connectivity, all of which remain expensive and constrained. Seed funding can support market expansion and software development, but scaling infrastructure globally may require further financing, supplier partnerships or customer-backed deployment structures.

The investment also reflects rising confidence in enterprise AI infrastructure as a distinct venture category. Unlike application-layer AI start-ups, companies such as Dapple are selling the rails on which production AI runs. That can create stickier customer relationships, particularly where deployments are tied to compliance, national boundaries and mission-critical workloads. It can also expose providers to slower sales cycles, demanding service-level obligations and competition from better-capitalised cloud incumbents.