The Vatican has just published a document of unprecedented scope for the technological world. The encyclical Magnifica Humanitas, written by Pope Leo, devotes more than 42,000 words to the ethical analysis of artificial intelligence. This text does not merely express wishful thinking; it directly tackles the economic and geopolitical structures that underpin the development of this technology. The sovereign pontiff denounces "the idolatry of profit" and warns against an unprecedented concentration of technological power in the hands of a few Silicon Valley giants, calling to "disarm" AI to place it at the service of human dignity.
The Technological Oligopoly Under Ethical Scrutiny
The concentration of power in the field of artificial intelligence is not just a financial issue: it is a question of cultural and cognitive sovereignty. Today, a handful of companies control the computing infrastructure required to train large language models (LLMs). According to an OECD report on the digital economy, this barrier to entry creates a de facto oligopoly. The consequences are manifold: the standardization of responses, cultural biases imposed on the rest of the world, and increased technological dependence for public institutions and the education sector.
The encyclical introduces the concept of "algor-ethics", an approach aimed at embedding moral principles at the very heart of algorithm design. The text insists that AI must not become a tool for cultural colonization or social exclusion. When public or educational decisions are delegated to proprietary systems hosted abroad, local communities lose their capacity for self-determination. To counter this drift, the Vatican advocates for decentralized governance, equitable access to technologies, and respect for privacy as fundamental rights.
This stance echoes previous initiatives, such as the Rome Call for AI Ethics, but it takes the reflection further by explicitly targeting business models based on the systematic capture of personal data. According to analyses published by the Financial Times and CNET, the encyclical marks a turning point by describing the excessive centralization of data as a threat to democracy and social peace.
Decentralization as an Architectural Response
It is precisely at this intersection of ethics, decentralization, and sovereignty that the design choices of the ProductivIA platform are situated. In contrast to the highly centralized model of major AI providers, the multi-silo architecture offers a concrete alternative. In this paradigm, data is not pooled in a single, opaque repository owned by a third party. Each organization, whether it is a school, a government department, or a business, has its own secure, isolated space, or "silo". This compartmentalization ensures that sensitive information remains under the exclusive control of the user, thereby meeting the requirements of Quebec's Law 25 on the protection of personal information.
To realize this independence, the platform integrates the sovereign Matania model. Unlike consumer services that systematically route queries to servers located abroad, Matania relies on models from the Qwen family hosted locally within Quebec. This approach allows sensitive data to be processed without compromising confidentiality, offering a viable alternative for organizations subject to strict compliance obligations.
Furthermore, to address the encyclical's call for equitable access and the protection of youth, applications like GoIA and ÉtudeIA have been designed with a philosophy of free access and transparency. GoIA offers a space for dialogue where users can compare responses from different models without ad tracking or the exploitation of their personal data. ÉtudeIA, for its part, is designed for schools, offering a virtual tutor that is grounded exclusively in the educational material provided by the teacher using retrieval-augmented generation (RAG) technology. This avoids the biases and commercial drifts often associated with unregulated consumer tools.
Going Further
The Magnifica Humanitas encyclical reminds us that technology is never neutral and that its technical architecture determines its moral scope. By prioritizing decentralization, respect for privacy, and local hosting, organizations can regain control of their digital destiny. The question is no longer simply whether we should use artificial intelligence, but how we can integrate it into a framework that respects human dignity and collective sovereignty.