Leo XIV’s encyclical on artificial intelligence comes at a unique time: one where algorithms are already deployed on the battlefields. Autonomous drones, targeting systems, lethal decision support – war is gradually being delegated to machines. Faced with this shift, the Pope makes a simple and radical affirmation: no decision involving human life can be entrusted to a machine, because it cannot bear moral responsibility; she can just fake it.
Moral responsibility requires a commitment to values, embodied through our bodies and our reasoning – certainly fallible and vulnerable, but Magnifica humanitas, according to the beautiful title of the encyclical. Leo XIV goes further than just the question of military AI. It reaffirms that no war can find moral legitimacy – thus rejecting the just war doctrine which had long served, and which still serves, as a theological cover for armed conflicts.
The same vision of AI, dominant and dominating
In this context, he explicitly deplores the rearmament of Europe and the discourses that accompany it, these stories of necessity which normalize, week after week, what would have scandalized the builders of international alliances in the decades after the Second World War. The common good, the intrinsic value of each human life, the primacy of dialogue and diplomacy – these values, he writes, exclude the logic of nuclear deterrence and the arms race.
Criticism of AI is part of a broader vision. What the Pope denounces is the culture of power which underlies both the technological war and the global race for AI: a logic of efficiency and profit which advances without taking into account its social and environmental costs, which concentrates wealth, widens inequalities, and threatens democracy through disinformation on an industrial scale. The Pope does not condemn AI as such – it is the vision behind it that is called into question: technocentric, utilitarian, indifferent to justice.
In other words, the same culture that produces autonomous weapons also produces surveillance algorithms, information bubbles and job casualization. They are the fruits of the same vision, dominant and dominating.
Combating the dizzying concentration of technological powers
This vision imposes itself under the guise of necessity, competitiveness, survival – as if resisting the race towards AI or rearmament was naivety or ignorance. However, the encyclical shatters this belief: it names what appears to be inevitable as being a political, economic and moral choice. She reminds us that we always have a choice.
It would be easy to criticize the Pope – and some have tried to do so without delay – by reducing him to a slayer of progress, out of step with his times, impervious to the challenges of economic innovation as well as to geopolitical realities. This would be, once again, defending a vision of history and technology by presenting it as the only possible one. It is urgent, on the contrary, to rely on the strong words of the Pope to promote other visions – both social and technological – and to remember that the future is not yet written: it is up to all of us, French citizens, European citizens, citizens of the world, to write it.
What can these alternative visions be? They begin, concretely, by informing citizens about how AI really works – beyond science fiction stories about the emergence of superhuman intelligence. AI remains a machine, certainly sophisticated, but a machine.
It is based on a value chain of dizzying concentration: a single company, Nvidia, supplies the chips necessary for all AI models in the world; these chips are manufactured by a single subcontractor, in Taiwan; the data centers that run these models belong to three American companies – Google, Amazon, Microsoft. Behind the promise of universal technology lies a deeply oligopolistic infrastructure.
When is artificial intelligence really useful?
We must therefore democratically open the debate on the real usefulness of AI. Do we need it, and what for? Are there models that consume less energy, and are designed for uses of general interest rather than for mass entertainment or the automation of business tasks? In which sectors can AI be deployed, under what conditions – regulation, testing, accountability – and with what objectives? Who is responsible for errors when algorithmic biases produce unfair decisions?
And above all: is it reasonable to invest massively in a technology whose environmental footprint – in energy, land use, water – is considerable, when the majority of its general public uses are leisure, and when there is a lack of resources to finance the climate transition?
Finally, we should shift the center of interest from the power of States – weapons, deterrence, technological competition – to the security of people: food, health, access to water, dignity. A paradigm shift which is not a utopia, but a political choice – one that our democracies still have the capacity to make.
About opinions
This text is signed by a guest author. He expresses his opinion and not that of the editorial staff. Our Live section aims to allow the expression of pluralism on religious, social and current affairs subjects, and to encourage dialogue, according to the criteria set by our editorial charter.
Share your opinion in comments or by writing to us at: readers.lacroix@groupebayard.com
