Berlin-based enterprise AI company Langdock has completed a corporate restructuring that runs against a familiar European start-up playbook: it has dismantled its US parent company and moved the group’s parent into a Germany-registered Societas Europaea (SE).
For many European companies, a US holding company is a practical route to venture funding, investor familiarity and access to startup networks. Langdock says that structure was initially important to its development, including its relationship with Y Combinator. But the company now sees the legal wrapper as a commercial complication rather than an advantage.
What changed
Langdock said it began the process in early 2026, spent several million euros on the restructuring and has now completed it. Its parent is now governed under European law through an SE structure registered in Germany.
The company says it serves roughly 13,000 organisations with a platform that provides access to multiple AI models, connects models to workplace data and applications, and supports agents and task automation. It reported an annual subscription revenue run rate of $50 million in August, up from $1 million in October 2024.
Its operating activity and customer data had remained in Germany even while the parent was US-registered. Still, Langdock said customers’ legal teams had to assess whether the US parent introduced legal or data-protection risk. The company says the US entity had no employees, infrastructure or access to production systems.
Why the corporate structure matters
For enterprise AI buyers, data residency is only one part of the procurement question. The jurisdiction of the vendor’s parent company can also trigger reviews about potential government access, contractual risk and governance. US laws including the Cloud Act have made those discussions more prominent, particularly for regulated customers and public-sector buyers.
Langdock is positioning a European parent as a way to simplify that assessment. Its argument is not that moving a holding company alone creates a sovereign technology stack. Rather, it removes one source of ambiguity for customers whose data and workflows may pass through AI systems.
That distinction matters. Enterprises evaluating AI platforms increasingly ask where data is stored, which entities control the service, which model providers are involved, and what legal regimes apply in a dispute or government request. A vendor that can answer those questions with a simpler corporate map may shorten sales cycles or face fewer objections.
A bet on European demand
Langdock says it wants to build a “sovereign, full-stack AI platform” over time. By year-end, it plans three new services and intends to begin operating its own German data centre for open-source models and computing capacity, scaling with customer demand.
The ambition is substantial, and the scale gap is equally clear. Langdock’s reported $50 million revenue run rate is rapid progress for a young company, but it remains small beside Amazon Web Services, which generated $128.7 billion in 2025 revenue, according to the company’s comparison.
Still, the company does not need to displace hyperscalers outright to make the strategy work. A more immediate opportunity is to serve organisations that want model choice, workflow integration and European governance without building their own AI platform.
What to watch next
The key test is whether legal and sovereignty positioning translates into durable operating advantages: faster procurement, stronger retention, or access to customers that would otherwise choose a larger US-based platform.
Investors will also watch whether an EU parent affects fundraising mechanics. Langdock says it still plans to raise internationally, while roughly 80% of the company is owned by founders and employees living in the EU.
For European founders, the move offers a useful counterexample to the assumption that a US holding company must be permanent. The cost and complexity of reversing that decision are real. But as AI procurement becomes more sensitive to jurisdiction and infrastructure control, company structure itself may become part of the product strategy.
