Enterprise AI adoption is running into a familiar problem: the more people an organization puts on AI tools, the harder usage-based pricing can become to manage. Cloudforce, the company behind nebulaONE®, is taking a different approach as it expands its National Harbor, Maryland headquarters by 15,000 square feet and adds 250 jobs. The move comes as the company’s AI platform surpasses 4 million users globally and raises a broader question about how enterprises will pay for AI at scale.

The enterprise AI market has spent much of the past two years debating models, agents and infrastructure. Increasingly, however, the economics of deployment are becoming just as important.

Cloudforce is betting that organizations want a different pricing model.

The AI platform company said it is doubling the footprint of its National Harbor headquarters with an additional 15,000 square feet, following support announced by Maryland Governor Wes Moore and the Maryland Department of Commerce for the company’s planned addition of 250 jobs.

The expansion comes as Cloudforce says its flagship nebulaONE platform now serves more than 4 million users worldwide.

The company’s pitch is not simply that more organizations want AI. It is that the traditional way of charging for enterprise software may become increasingly difficult to reconcile with mass AI adoption.

Moving beyond per-seat AI pricing

Most enterprise software has historically been sold on a per-user or per-seat basis.

AI introduces a complication. Many AI services involve variable inference costs, with usage often linked to tokens, model size, context length and frequency of interaction. As organizations attempt to make AI available to entire workforces, those costs can become difficult to forecast.

Cloudforce says nebulaONE instead uses a value-based licensing model, allowing institutions to provide access to all users without purchasing individual AI seats.

The company claims the approach has reduced AI adoption costs by roughly an order of magnitude.

That figure is a company claim rather than an independently verified industry benchmark, but the underlying issue is real: AI pricing is becoming a strategic consideration for CIOs and procurement teams.

The market is moving toward a mixture of pricing models. Some vendors charge per seat, others by usage, tokens, API calls or compute, while enterprise platforms increasingly combine subscriptions with consumption-based services.

For buyers, the question is shifting from How many employees need an AI license? to What will widespread AI usage cost at enterprise scale?

Private AI infrastructure is another differentiator

Pricing is only part of Cloudforce’s strategy.

nebulaONE is designed to give organizations access to frontier AI models and agentic workflows within their own cloud environments. Cloudforce says customer data remains under the organization’s control and is excluded from model training.

That architecture targets one of the biggest enterprise barriers to generative AI: data governance.

Organizations in education, healthcare, financial services and government often have significant restrictions around sensitive information. Sending data into an external consumer AI service may therefore be unacceptable even when the underlying model is capable.

A private or customer-controlled deployment can address some of those concerns, although enterprises still need to evaluate identity management, encryption, model-provider contracts, auditability, data residency and agent permissions.

Cloudforce is effectively positioning nebulaONE between general-purpose AI services and fully custom enterprise AI infrastructure.

Education is becoming an early proving ground

Cloudforce’s strongest disclosed traction appears to be in education.

The company says Microsoft named it Global Education Partner of the Year, citing nebulaONE deployments at institutions including the University of Maryland, UCLA and the University of Oxford.

Education is an interesting market for enterprise AI because institutions have large and diverse user populations. A university may need to support students, faculty, researchers and administrative staff, making per-seat economics particularly consequential.

It is also a sector with substantial privacy and governance requirements.

That combination creates a useful test for the broader proposition that institutions can deploy AI broadly without allowing costs or data-control concerns to prevent adoption.

Capital is following the expansion

Cloudforce’s physical expansion follows a $10 million Series A led by Owl Ventures and M12, Microsoft’s venture fund.

The company says its workforce has grown 83% over two years and is approaching 150 employees. Its National Harbor expansion will therefore materially increase its local footprint.

Maryland and Prince George’s County are supporting the project through conditional loans from Advantage Maryland and the Prince George’s County Economic Development Corporation, which Cloudforce says will help cover construction and materials.

The company’s decision to remain in National Harbor also fits into a broader effort by Maryland to build a technology and innovation economy around the Washington metropolitan region.

For Cloudforce, the financing structure has another advantage: more corporate capital can remain available for product development and hiring rather than being consumed by the physical cost of expansion.

The bigger competition is platform consolidation

Cloudforce is entering an increasingly crowded enterprise AI market.

Microsoft, Google, Amazon, OpenAI, Anthropic, Salesforce and NVIDIA are all expanding their enterprise AI ecosystems, while a growing number of infrastructure and orchestration companies are competing to help organizations deploy models securely.

The competitive challenge for smaller AI platforms is therefore significant.

Owning the customer relationship requires more than access to a frontier model. Model capabilities are increasingly commoditized through APIs, while cloud providers can bundle AI into existing enterprise contracts.

Companies such as Cloudforce need to differentiate through deployment architecture, governance, workflow integration, pricing and domain-specific functionality.

Its value-based pricing model is one such differentiator.

The more interesting test will be whether that model can remain economically sustainable as AI inference becomes cheaper, models become more efficient and enterprise workloads become increasingly agentic.

AI adoption is entering an economic phase

Cloudforce’s expansion illustrates a broader transition in enterprise AI.

The first wave was about convincing companies that generative AI could be useful. The second is about making it secure, manageable and economically viable for millions of users.

That means pricing architecture matters almost as much as model capability.

If enterprises truly want AI available across entire organizations, per-seat licensing may become less attractive in some environments. Conversely, consumption-based pricing can expose customers to unpredictable costs as usage grows.

There is unlikely to be one winning model.

But the companies that can make AI simultaneously accessible, governed and financially predictable may have an advantage as enterprises move from experimentation to ubiquitous deployment.

Cloudforce’s 4-million-user milestone and Maryland expansion are an early signal of that shift. The next test is whether its economics can scale as quickly as its user base.

Market Landscape

Enterprise AI infrastructure is separating into several layers: foundation models, cloud infrastructure, AI orchestration, security and governance, and end-user applications.

Cloudforce competes primarily in the orchestration and enterprise deployment layer, while leveraging models and infrastructure from the broader AI ecosystem.

Its emphasis on private cloud deployment reflects a growing enterprise requirement for data sovereignty, governance and controlled access to AI agents. Its pricing strategy addresses another emerging concern: unpredictable AI consumption costs.

The competitive environment is intensifying. Microsoft, Google and Amazon can integrate AI into existing cloud and productivity ecosystems, while specialist providers can compete through flexibility and multivendor support.

For CIOs, the important comparison is therefore not simply model quality. Buyers increasingly need to evaluate total cost of ownership, data controls, model portability, governance, agent security and pricing predictability.

Top Insights

• Cloudforce is expanding its Maryland headquarters by 15,000 square feet while adding 250 jobs as nebulaONE surpasses four million global users.
• The company’s value-based licensing model challenges per-seat AI pricing, addressing enterprise concerns about unpredictable token costs as AI access expands across workforces.
• nebulaONE provides private access to frontier models and agentic workflows within customer cloud environments, targeting organizations with demanding data governance requirements.
• Cloudforce’s education deployments highlight a market where large user populations, privacy requirements and institutional budgets make enterprise AI economics particularly important.
• The company faces competition from Microsoft, Google, Amazon and specialist AI platforms, making pricing, governance and multivendor model access key differentiators.

Power Tomorrow’s Intelligence — Build It with TechEdgeAI