Article translated from French and originally published in The Conversation.
Author: Lucas Eustache, Doctorant à la Chaire Gouvernance et Regulation, Université Paris Dauphine – PSL
The liquidation of Agdatahub raises questions about the legal and economic status of data platforms designed for a specific profession. While the initial idea was to give farmers back control over their data, the solution highlighted the chosen approach’s shortcomings. Data platforms are not just technical in nature. Their political and strategic dimensions must also be taken into account.
As data becomes a strategic resource, its governance is emerging as a matter of sovereignty. In agriculture, connected equipment collects crucial data, which is often stored by agricultural equipment manufacturers without farmers having full access to it. This technical dependence undermines their economic autonomy.
Agriculture is undergoing a digital revolution that is still largely driven by industry. In a data-driven digital economy, the ability of public actors to organise a fair, transparent and sustainable digital space is a key variable.
Agdatahub: a textbook case
The experience of Agdatahub provides a concrete illustration of these challenges. Launched in 2020, this platform aimed to structure agricultural data sharing (plots, weather, equipment, farming practices) between producers, manufacturers and their multiple service providers.
It offered a secure portal, based on European standards, enabling detailed consent management and system interoperability. Supported from the outset by the State through France 2030, the Banque des Territoires and the Ministry of Agriculture, Agdatahub was recognised as a strategic player in the sector’s digital roadmap.
In 2023, it coordinated the European project Agridataspace, which aims to create a future common agricultural data space to make digital sovereignty issues operational in this sector.
Until now, much of the value generated by data has been captured by agricultural machinery manufacturers, particularly in the United States. These companies offer integrated solutions that combine machinery and automated data collection and processing. Already widely adopted in the field, these offerings lock in access to data. This is to the detriment of farmers, who lose control of their own information and the ability to extract value from it themselves.
Support from the authorities was not enough
Despite support from the French and European authorities, Agdatahub was liquidated at the end of 2024. Several factors explain Agdatahub’s failure.
Concrete use cases remained limited. Beyond consent management, few applications really found their place in operators’ practices. The economic incentives to share or monetise data via the platform were weak, both for producers and manufacturers, in the absence of immediate benefits.
At the same time, the platform had to contend with competition from integrated private players with an already well-established service offering. Public support, although decisive at the time of launch, was not extended over time, weakening the project’s financial structure.
Finally, the government did not fully support the platform’s momentum, keeping certain strategic public datasets closed, thereby limiting the possibility of developing high value-added services, such as parcel data (telepac data) or access to data from the agricultural asset register (now the national business register), which would have enabled an automatic identity verification mechanism on the platform.
In 2025, a French free software publisher took over the activity with the ambition of building a more open, cooperative and sustainable governance model.
Long-term solutions
This trajectory reminds us that data sharing cannot rely solely on tools or intentions, but requires the implementation of genuine socio-technical solutions that are designed for the long term and supported by all stakeholders.
A data sharing ecosystem can be defined as a system in which various stakeholders, companies, local authorities, researchers, experts, etc., collaborate on data exchange. The ultimate goal is to develop use cases according to specific rules. Unlike a single centralised platform, these ecosystems rely on the diversity of stakeholders: data producers, re-users, hosts, intermediaries and the existence of a trusted third party that guarantees value sharing and member control over their data.
The example of Agdatahub clearly illustrates this. Circulating data is not just a technical or regulatory issue. It involves combining robust technological standards with sustainable forms of collective commitment, without which sharing infrastructures will remain nothing more than a promise. Even when the technical building blocks are in place (consent management, interoperability, secure data flows), the ecosystem will only work if the players find mutual incentives and recognise each other as legitimate partners.
Weak collective commitment
The partial failure of Agdatahub reveals precisely this imbalance: solid technical foundations, but weak collective commitment due to limited economic incentives and insufficient public commitment over time. This case highlights that even a well-designed technical infrastructure cannot function without shared governance and aligning mechanisms between actors.
From this perspective, the state appears as a key actor: not to direct the ecosystem but to guarantee its conditions of existence by ensuring a stable framework, pooling certain initial costs, and opening up key public data sets. It is in this articulation between technical infrastructure, collective commitments and public regulation that true shared information sovereignty can be built.
Towards a status as infrastructure of general interest?
The example of Agdatahub raises a key public policy question: should these ecosystems be given specific legal status? Considering data, its structures and its usefulness within a market as essential infrastructure is not new. This framework would support hybrid structures that are both public and private, but with an explicit digital public service mission. This would promote economies of scale, convergence of standards and pooling of management costs, while ensuring transparent governance.
The state could play a role similar to that which it plays in other strategic sectors such as energy, public transport and health: guaranteeing equitable access to resources, supporting technological and organisational transitions, and ensuring the economic and social sustainability of models.
Sharing data is not simply a matter of opening up access to files. It involves organising a genuine shared infrastructure based on standards, governance mechanisms and pooling arrangements. In the agricultural sector, this infrastructure plays a key role in reducing inequalities in access to digital tools and strategic information. In the agricultural world, small and medium-sized farms and local cooperatives rarely have the resources to develop their own data processing or cross-referencing solutions. By facilitating access to data from remote sensing, weather sensors and public policies (subsidies, zoning, risks), a shared data infrastructure reduces their dependence on actors who concentrate these data and technologies, such as agricultural machine operators, input suppliers and other equipment manufacturers.
In concrete terms, this can translate into better anticipation of climate risks, optimised agricultural practices and more active participation in traceability and certification schemes. The challenge is both economic and social: by ensuring that the digital transition does not only benefit the best-equipped players, data infrastructure enables a rebalancing of relationships within the agricultural value chain.
Like roads or energy networks, this infrastructure requires robust, balanced and stable governance over time. The example of Agdatahub shows that even a technically sound and politically supported project can fail without a sustainable framework.
At a time of digital and ecological transition, rethinking the state’s role in the data economy is becoming essential to ensure that the conditions for trust, fairness and the common good are met.
