The next wave of enterprise AI may not arrive as another software platform that companies simply subscribe to.
For industrial businesses, some of the most valuable AI applications sit much closer to physical operations, including factories, vehicles, logistics networks and machinery. Those systems depend heavily on proprietary operational data and can become important enough that enterprises may not want the same technology available to competitors.
Vantora, formerly known as UP.Labs, is building its business around that idea. The company has secured more than $100 million from Silversmith Capital Partners, its first outside investment since launching in 2022, as it expands a model for creating AI-native companies alongside major industrial enterprises.
The investment will support additional corporate partnerships, development of Vantora’s data ontology technology and hiring across AI and commercial roles. But the funding comes alongside an equally important change in how the company approaches enterprise innovation.
From Startup Studio to Enterprise AI Builder
Vantora does not operate like a traditional venture capital firm or accelerator.
Its teams work directly inside large companies to identify operational problems with significant financial value. Vantora then brings together founders, product specialists and AI engineers with the enterprise’s own operators and data to build a new company around the problem.
According to the company, the opportunities it targets typically have the potential to contribute between $50 million and $100 million in annual EBITDA to a corporate partner. Vantora has worked across industries including automotive, aviation, transportation, manufacturing, energy and retail.
Its existing corporate relationships include Porsche, Alaska Airlines, J.B. Hunt, Wabash and TDG, the parent company of Ashley Furniture. The company has also expanded into industrial manufacturing and oil and gas, although it has not disclosed the names of some newer partners.
The model has already produced 17 ventures, according to reporting from The Wall Street Journal, with Vantora aiming to reach 20 by the end of this year. The company is also profitable and has increased revenue by 79% over the past year.
Why Vantora Is Moving Deeper Into Physical AI
Vantora’s strategy has evolved as enterprises have become more protective of the technology being developed around their most important operations.
Under its earlier UP.Labs model, some startups could eventually address a broader market beyond the original corporate partner. That approach created a problem when an AI system was closely connected to a company’s proprietary machinery, processes or competitive advantage.
Founder and CEO John Kuolt told TechCrunch that Vantora had previously abandoned some promising ideas because its corporate partners considered the technology too strategically sensitive to offer externally.
The company’s new model changes that equation.
Corporate partners can invest in the ventures, become their first customers and ultimately have the option to bring those businesses into their own organizations. Kuolt describes the structure as creating a proprietary M&A pipeline.
That structure is especially relevant to physical AI, where artificial intelligence moves beyond software and becomes connected to machines and physical operations.
An industrial company trying to make its equipment more autonomous, for example, may see the intelligence controlling those machines as proprietary infrastructure rather than software it wants shared across an industry. Vantora can build the technology around that specific environment while giving the enterprise a path to retain ownership.
Giving Enterprises Ownership of the AI Layer
The approach represents a different answer to a question many large companies face when adopting AI: should they buy technology from an external vendor, build it internally or create something between the two?
Vantora is effectively offering the third option.
The enterprise contributes its operational expertise, data and real-world problem. Vantora provides the entrepreneurial and technical teams needed to turn that opportunity into a functioning AI company.
For companies with highly specialized operations, that can address one of the limitations of general-purpose enterprise AI. A model or software platform available to thousands of customers may improve productivity, but it does not necessarily capture the processes that make one manufacturer’s factory or one logistics company’s network different from another.
Physical AI makes that distinction more important because the technology can directly influence how equipment and infrastructure operate.
Vantora’s work with J.B. Hunt illustrates the issue. Kuolt said the firm previously identified an AI opportunity with the transportation company but did not pursue it because the technology was considered too strategically important to take to the broader market. Under the new model, projects like that can remain proprietary to the corporate partner.
A Different Route to Enterprise AI Adoption
Silversmith’s investment gives Vantora more resources to expand that model at a time when industrial companies are trying to determine where AI can create measurable operational value.
The funding is notable because Vantora had remained founder-led and profitable without institutional capital. Silversmith, which manages more than $5 billion in capital, is now its first institutional investor.
Vantora is also developing its own data ontology product and plans to expand its AI and commercial teams as it adds more corporate partnerships.
For enterprises, the model changes the usual relationship with an AI startup. Instead of waiting for an external vendor to develop a product and then adapting it to their operations, the company becomes involved from the beginning, using its own data and domain expertise to shape the technology.
That could prove particularly useful in industries where the most valuable AI applications are tied to proprietary equipment, workflows and operational knowledge. Vantora’s bet is that some companies will want more than access to those systems. They will want to own them.