AI transparency

Philosophy Arena is built on artificial intelligence, and this page states plainly where you encounter it.

Every conversation — Dialogue, Debate, Sophia, and the Curator — is generated by an AI language model. The philosopher voices are AI renderings grounded in the thinkers' actual texts; they are not the philosophers, and their answers can be wrong. Cited passages come from the ingested works themselves.

The philosopher and work pages carry written introductions — summaries, teasers and context notes. These are written by AI and fact-checked against reference sources. They are written to interest you in a thinker or a book rather than to read as a neutral reference, so they carry editorial judgement: the dates, names and publication facts are checked, the characterisation is a point of view. German versions of these pages are translated from the English by AI and edited. Translations of cited source passages may be machine-generated and are labeled as such where they appear.

Every philosopher portrait on this site is an AI-generated rendering. None of them is a photograph, a painting, or a reproduction of a historical artwork, and none is a likeness checked against a historical source. They appear wherever a philosopher is shown: the philosopher index, the individual philosopher pages, the philosopher of the week in the Atrium, the medallions beside replies in a conversation, and the preview cards this site produces when a page is shared.

They are deliberately rendered as two-tone engraved medallions rather than lifelike faces. The treatment is the disclosure: it is meant to read as an illustration of a thinker at a glance, in the way a frontispiece or a coin does, and never as evidence of what that person looked like.

This disclosure will be extended as the product grows (Art. 50 EU AI Act).