Cohere Acquires Aleph Alpha in $20B Deal to Build a Transatlantic Sovereign AI Powerhouse

In a deal announced Thursday in Berlin, with representatives from two governments standing at the podium alongside the CEOs, Cohere has agreed to acquire Germany’s Aleph Alpha — creating a combined entity valued at approximately $20 billion that both companies are positioning as the world’s premier sovereign AI provider.

The merger brings together Cohere, the Toronto-based enterprise AI model company co-founded by Google Brain veteran Aidan Gomez, with Aleph Alpha, the Heidelberg startup that has spent the better part of three years building AI infrastructure for European governments and regulated industries. The combined company will be headquartered in Toronto and maintain significant European operations out of Germany.

Structure and Financing

Under the deal terms, Cohere’s shareholders will receive approximately 90% of the combined company, with Aleph Alpha’s shareholders receiving the remaining 10%. A concurrent Series E funding round, which has not yet formally closed, is expected to cement the valuation at $20 billion.

Leading the Series E is Schwarz Group, the German retail conglomerate that owns Lidl and Kaufland, which is investing $600 million. Schwarz had previously been an investor in Aleph Alpha and has been expanding its enterprise technology ambitions through its IT subsidiary Schwarz Digits. The size of that investment — $600 million from a single anchor investor — speaks to the confidence Europe’s business establishment has in the sovereign AI thesis.

Other investors in the round were not disclosed at announcement time, though sources familiar with the deal indicated that additional European strategic investors are expected to participate before the round closes.

The Sovereign AI Thesis

The phrase “sovereign AI” has become something of a buzzword in European policy circles, but the underlying commercial reality is concrete: a growing number of governments, defense agencies, healthcare systems, and financial institutions require AI infrastructure that does not route data through or depend upon American or Chinese cloud providers.

For European public sector customers, this has historically meant painful trade-offs — either accepting the legal and political risks of U.S.-hyperscaler dependence or making do with less capable domestically-built alternatives. Aleph Alpha has spent years trying to thread this needle, building GDPR-compliant infrastructure and winning contracts with German federal agencies, the European Space Agency, and several large German industrial companies.

Cohere brings what Aleph Alpha lacked: frontier-grade model capability, a proven enterprise sales motion, and a global customer base that includes major names in finance, healthcare, and technology. Cohere’s Command R and Command R+ models have become a quiet staple of enterprise retrieval-augmented generation (RAG) pipelines, prized for their reliability and commercial licensing terms.

Together, the argument goes, the combined company can offer something no other AI vendor can: genuinely sovereign deployment (on-premises, private cloud, or sovereign cloud infrastructure in the jurisdiction of the customer’s choosing) backed by a model that is actually competitive with OpenAI and Anthropic.

A Geopolitical Deal as Much as a Commercial One

The announcement ceremony was notably staged in Berlin, not Toronto, and both Germany’s Federal Minister for Digital Affairs, Karsten Wildberger, and Canada’s Minister for AI and Digital Innovation, Evan Solomon, were present. That degree of government participation in a private M&A announcement is unusual and intentional.

Canada and Germany are both navigating the same structural anxiety: dependence on U.S. AI infrastructure at a moment when the policy environment in Washington has become volatile and unpredictable. The tariff wars of 2025, the unpredictable posture of the current U.S. administration toward traditional allies, and the broader decoupling dynamic have all accelerated European and Canadian interest in building independent AI capabilities.

Solomon framed the deal in explicitly geopolitical terms, describing it as “a statement that democratic countries can build world-class AI on our own terms.” Wildberger called the merger “a European success story with global ambitions.”

Competitive Implications

The combined Cohere-Aleph Alpha entity enters the market at a moment of significant AI industry consolidation. The two companies most directly in its crosshairs are OpenAI and Anthropic, both of which have aggressively expanded their enterprise sales organizations and are bidding on government contracts across Europe and Canada.

Cohere CEO Aidan Gomez has consistently argued that enterprises are increasingly wary of being locked into a single U.S. AI provider — and that concern has only deepened as OpenAI moves toward a for-profit structure and a potential IPO that could prioritize shareholder returns over enterprise stability commitments. That argument will become considerably more persuasive with Aleph Alpha’s sovereign infrastructure credentials and European government relationships behind it.

The deal also has implications for the broader enterprise AI landscape. AI21 Labs, Mistral, and other “alternative” enterprise AI providers will find the competitive space somewhat narrowed — the combined company will have more capital, more talent, and more customer relationships than any of them individually.

Aleph Alpha’s Trajectory

Aleph Alpha was founded in 2019 by Jonas Andrulis, who had previously worked at Apple’s AI labs. The company raised €500 million in 2023 in one of the largest European AI rounds of that year, at a valuation of approximately $5 billion. The 10% share in the combined $20 billion entity implies Aleph Alpha shareholders are receiving a valuation of roughly $2 billion — a markdown from the 2023 funding round, which reflects the reality that Aleph Alpha, for all its government relationships, never achieved the commercial model velocity to justify its prior valuation independently.

Andrulis will take a senior role in the combined company, focused on European government and regulated industry business — the domain where his network is deepest.

What’s Next

The Series E is expected to close within the next several weeks, with full regulatory approvals anticipated by end of Q2 2026, given that the deal does not obviously trigger major antitrust concerns in either jurisdiction.

For enterprise technology buyers, particularly those in regulated European industries who have been watching the sovereign AI space, the deal may finally provide the answer they have been waiting for: a single, capable, contractually reliable alternative to the U.S. hyperscaler-AI stack. Whether Cohere and Aleph Alpha can execute on that promise at scale — building and deploying competitive models while also managing the complexity of sovereign infrastructure across multiple jurisdictions — is the central question the combined company will need to answer.

The market, at least, has given them a $20 billion vote of confidence to try.