Write into the statutes protective clauses against a non-EU acquisition — shareholders’ agreement, European pre-emption rights, golden share, specific share with blocking rights#

What this is, concretely#

This commitment addresses the capital lock that is missing from most European publishers. Without a statutory lock, a French publisher changes its reference jurisdiction overnight: it suffices for a non-EU acquirer to buy a majority of the capital, and the entire sovereignty grid — chain of decision, applicable jurisdiction, extraterritorial exposure — flips. Recent cases bring it home starkly: Aleph Alpha taken over by Cohere in April 2026, Silo AI by AMD in August 2024, MariaDB by K1 Investment in September 2024. None of these actors had a blocking statutory clause.

The commitment consists of writing into the company’s statutes, or into a publicly referenced enforceable shareholders’ agreement, protective clauses that can be activated in the event of an attempt at non-EU takeover. Several mechanisms coexist:

• Shareholders’ agreement restricting block transfers above a threshold to entities whose ultimate control is European, with sanctions in case of breach.
• European pre-emption rights for the benefit of an identified European fund, a public institution, or a consortium of European shareholders, in case of a transfer above a threshold.
• Golden share or specific share with blocking rights, held by a European public or para-public entity (the model of framework-controlled privatisations in several Member States), with veto rights on strategic operations.
• Change-of-control clauses in client contracts, which trigger a free termination right in case of non-EU takeover.

These arrangements are distinct from but complementary to state foreign-investment screening frameworks (IEF in France, AWG in Germany, Golden Power in Italy), which act downstream and only on actors deemed strategic by the state. The statutory lock, by contrast, is raised by the publisher itself, for its own clients, regardless of whatever attention the state may choose to give it.

Why this commitment matters#

Thesis 13 of the manifesto grounds this commitment: “Technological sovereignty is not an end in itself. It is the condition under which an organisation — a business, an administration, an individual — can continue to operate its data and conduct its operations, whatever happens: provider failure, geopolitical conflict, sanctions, hostile takeover, unilateral flip by a publisher. It is a right, not a comfort.” The explicit mention of “hostile takeover” in this thesis is no afterthought: it is one of the five scenarios against which sovereignty must hold, and it is precisely the one that contractual commitments (reversibility, notice period, escrow) do not cover, because a non-EU acquisition can redefine the publisher’s jurisdiction without touching its contracts.

This commitment forms part of the new section “Conditions of equivalence for the proprietary publisher” of the About page. The “anti non-EU acquisition statutory clauses” condition is the sixth and most structural: it acts upstream of contractual relationships, on the capital chain itself. It complements pub-001-publish-sovereignty-profile (the Profile documents the current jurisdiction) by guaranteeing that this documented jurisdiction cannot be modified by an abrupt and unforeseen capital event. It also complements pub-005-establish-software-escrow and pub-011-operational-continuity-arrangement by preventing a change of control from emptying these arrangements of their meaning (a non-EU acquirer could in theory repudiate an escrow set up under French law, or even invoke a duty of cooperation with a foreign jurisdiction).

A concrete example#

A French publisher of generative AI specialised for regulated markets, around 90 employees, capital structure including two French funds and one German fund, takes this commitment in May 2026 with a 24-month horizon. The operation is conducted in several stages. First half of 2026: legal audit of the existing statutes by a firm specialised in company law and European law; identification of the levers compatible with the existing shareholders’ agreement. Second half of 2026: drafting of a new shareholders’ agreement integrating a pre-emption right for the benefit of a consortium of existing European shareholders for any transfer above 10% of capital, and a change-of-control clause triggering free termination rights in all client contracts in case of non-EU takeover.

First half of 2027: approach to Bpifrance and a European sovereign fund to structure a specific share with blocking rights covering transfer, demerger, and strategic-asset transfer operations (training datasets, model weights, proprietary code). Second half of 2027: signing, publication of the new statutory provisions (excerpts, while respecting the confidentiality of commercial counterparties), explicit mention in the Sovereignty Profile. By the horizon, two public-sector clients and one operator of vital importance cite the robustness of this architecture as a determining factor in their long-term commitment.

Anti-pattern to avoid#

A declaration of good intent (“our management is not contemplating any transfer to a non-EU acquirer”) without an enforceable statutory lock has no legal value. A shareholders’ agreement that is not publicly referenced prevents clients from relying on it. A pre-emption right for the benefit of a fund that does not actually have the means to exercise it is window-dressing. A golden share entrusted to a foreign entity, even an allied one, empties the arrangement of its meaning. Finally, conflating the statutory lock with IEF screening (which depends on strategic qualification by the state, and acts at the moment of the transaction without preventing every case) amounts to relying on a mechanism whose triggering you do not control.

Success indicators#

By the 24-month horizon, you can reasonably consider this commitment fulfilled if the statutes or shareholders’ agreement integrate at least two of the mechanisms described, if these provisions are publicly referenced (excerpts, mention in the Sovereignty Profile), if they have been validated by an independent legal firm, and if critical clients have received explicit communication about the robustness of the capital lock put in place.

JSON schema category: publication. Default horizon: 24 months. Applicable to: businesses.