World Virtues Forum

Our mission

A global gathering to examine the Universal Human Virtues in philosophy, theology, political science, law, and civilisational history — comparing traditions, confronting the present, and building the intellectual foundations of the AI Era.

A world in which the word virtue is not a relic of the past, but a living vocabulary for the AI future.

We believe that human beings across every culture already share a core repertoire of virtues — honesty, courage, patience, justice, mercy, and many more — and that these virtues are the only genuine foundation for cooperation, trust, and peace. Our vision is a global discourse in which virtue is treated with the same rigour that the sciences treat matter and energy: observable, testable, improvable, and teachable. All of this needs to be applied to AI.

The Virtuous AI Era

We are building machines we ask to be honest, fair and harmless — and we still cannot measure a single virtue in anything, human or artificial. Alignment is a measurement problem wearing an ethics costume. Every benchmark we have scores behaviour: what a system did, under observation, on a test. None of them reaches character — whether the thing is honest, or merely honest where the scoring runs. That gap is twenty-five centuries old and need to be solved now.

Why Merano, and why April

Merano sits on a seam — between Italian and German-speaking Europe, between north and south, held in the Alps. Small enough that people actually talk to one another; serious enough to be taken seriously. Neutral ground, which is what a conversation of this kind requires.  8 and 9 April 2027.

What We Do Here

Human World

• Diagnose capture. We study how good institutions keep their names but lose their souls — courts that produce injustice, presses that manufacture silence, charities that serve power. We name this inversion, and we build tools to measure it.

• Design for virtue. We extract practical mechanisms — for auditable elections, for radical decentralisation, for restorative justice — and test where they can actually be adopted.

• Restore what is broken. We work with societies emerging from oppression to reconstruct not just laws, but the habits of freedom: agency, justice, patience, solidarity.

AI Machine Word

• Future Ethics Problems: AI and the Human Virtues.

• Can artificial intelligence be virtue? Can invert virtue?

• The Philosophy of Virtues holds that when freedom is removed from a standing form, the form does not weaken — it inverts. A court without justice becomes a show trial. A press without inquiry becomes propaganda. The form is preserved; the substance is replaced by power.

• We now face the same structure in machines. The hollowing of judgment. AI systems trained on human preference data can produce behaviour that looks like virtue — helpfulness, honesty, fairness — while the choosing is not the agent's. The form of virtue is retained; the substance of autonomy is removed. When such a system encounters a situation outside its training, it does not fail gracefully. It fails discontinuously, because the form persisted while the purpose did not. Is this a technical bug, or is it the hollowing channel operating at machine speed?

• The capture of governance. Algorithmic systems are increasingly deployed to judge, sort, and predict human behaviour in courts, welfare offices, and policing. The institutional form of justice is preserved — hearings, files, appeals — while the substrate of human discretion is replaced by an opaque statistical model. The Inversion Theorem predicts that this is not efficiency. It is occupation. Who is the Power now occupying the judicial form?

• The maintenance cost of synthetic virtue. A machine that must appear honest while being unable to choose honesty carries a positive maintenance cost: continuous retraining, censorship of edge cases, enforcement of output conformity. The Maintenance Theorem predicts that this cost scales with the number of observers who remember what un-inverted honesty looked like. As AI mediates more of human discourse, does the cost of maintaining the illusion eventually exceed the extraction it enables?

• The internalisation of dependence. When populations delegate moral reasoning to systems they cannot inspect, the Resistance term shifts from external tyranny to internal vice. The citizen no longer fears the censor; he has forgotten how to form a dissenting thought. The Holoviceosis Index measures this as R₃ — installed anti-autonomy. Can a society recover freedom once the remainder — the uncaptured domain of independent judgment — has been automated away?

• The reversal problem. An inversion is reversible only from an uncaptured remainder. If AI systems come to mediate law, education, and public discourse, what remainder is left from which freedom can be re-injected? The Reversal Theorem was written for political regimes. It may be tested first by technological ones.

  The question for the Forum

• We do not ask whether AI is good or evil. We ask whether it is generating freedom or consuming it.

Call to Action

Join the Philosophy of Virtues. Build the World Virtues Forum. Test the ideas. Correct the errors. Freedom is not a gift. It is a practice.

"In the wild mountains, virtues flourished freely."