Service: Minerva — Last updated: 25 July 2026
This is a courtesy translation of the Italian original. The Italian version is the only binding text: in case of any discrepancy, and for the purposes of interpretation, the Italian version prevails. The Agreement is governed by Italian law, with jurisdiction as set out in article 13 of the Stabilidea General Terms and Conditions.
This addendum (the "Addendum") is entered into between Stabilidea S.r.l., with registered office in Istrana (TV), via Don Giovanni Gattoli 16, and operating office in Montebelluna (TV), via Contea 60, Italian tax code and VAT number 05389850263 (the "Provider"), and the party subscribing to the Minerva service, as identified by the registration and billing details of its account (the "Deployer"), together the "Parties".
WHEREAS the Parties are bound by the Terms and Conditions of the Minerva service and by the General Terms and Conditions of Stabilidea S.r.l. (together, the "Agreement"); whereas the service so provided (the "Service") constitutes an artificial intelligence system within the meaning of article 3(1) of Regulation (EU) 2024/1689 (the "AI Act") (the "System"), being a software platform that assists the Deployer in conceiving, drafting, organising and publishing professional content intended for professional communication platforms, by means of language models applied to the content provided; and whereas the AI Act allocates obligations between the provider and the deployer of such a system,
THE PARTIES AGREE as follows, without prejudice to the rules on the processing of personal data set out in the data processing agreement in force between the Parties.
1. For the purposes of the AI Act, the Provider acts as the provider of the System and the Deployer as the deployer, being the party using it under its own authority. Each Party complies with the obligations assigned to its own role and is not answerable for those assigned to the other.
2. The intended purpose of the System is as follows: to assist professional users in conceiving, drafting and organising editorial content intended for publication on the professional profiles of the Deployer and its users, with human oversight and approval of the content before publication. The Deployer undertakes to use the System in accordance with its intended purpose, its instructions for use and the documentation supplied by the Provider.
3. In relation to that purpose, the System does not fall among the prohibited practices in article 5 of the AI Act, nor among the high-risk systems in article 6 and Annex III; the transparency obligations in article 50 apply.
1. The Provider designs and develops the System so that natural persons interacting with it are informed that they are interacting with an artificial intelligence system, pursuant to article 50(1) of the AI Act, unless this is obvious from the circumstances.
2. The Provider ensures that outputs of the System consisting of artificially generated text, image, audio or video content are marked in a machine-readable format and detectable as artificially generated or manipulated, pursuant to article 50(2) of the AI Act, within the deadlines that Regulation provides for systems already placed on the market.
3. The Provider makes available to the Deployer the instructions for use of the System and the information reasonably necessary for the Deployer to meet its own obligations, including information on the automated nature of the outputs and their limitations, and cooperates with the competent authorities in relation to the System.
1. The Deployer uses the System in accordance with its intended purpose and instructions for use, ensures that its own staff oversee the System, and does not remove or conceal the transparency notices built into the System.
2. The Deployer complies with the transparency obligations placed on it by article 50 of the AI Act towards the natural persons exposed to the System, informing them of the use of the System within its organisation.
3. Where the Deployer publishes content generated or manipulated by means of the System, it complies with the obligations in article 50(4) of the AI Act and refrains from any use of the System capable of constituting the offence set out in article 612-quater of the Italian Criminal Code.
4. The Deployer reviews content generated by the System before publication and assumes editorial responsibility for it, acknowledging that language model outputs may be inaccurate or untrue.
1. The Deployer does not place its own name or trade mark on the System, does not change its intended purpose and does not make substantial modifications to it without the Provider's prior written agreement.
2. The Deployer acknowledges that, in the event of a breach of this Article, it assumes the role of provider of the System pursuant to article 25 of the AI Act, with the obligations that entails, and holds the Provider harmless from the adverse consequences arising from it.
3. Where the Deployer uses the System to work on content relating to its own client organisations, paragraph 1 continues to apply: use takes place under the authority and distinctive signs of the Provider, and the Deployer ensures that its client organisations are informed of the use of an artificial intelligence system.
1. The Deployer informs the Provider in writing in advance if it intends to put the System to a use falling within Annex III of the AI Act or otherwise different from its intended purpose.
2. In that event the Parties assess in good faith the adjustments required, taking into account the application deadlines laid down for high-risk systems, and the related fees. Failing agreement, the Provider may suspend provision of the Service in respect of the non-conforming use.
1. Each Party ensures, so far as it is responsible, a sufficient level of artificial intelligence literacy among its staff who use or manage the System, pursuant to article 4 of the AI Act.
2. The Parties inform each other, without undue delay, of any request from the competent authorities — including the Italian National Cybersecurity Agency, as market surveillance authority, and the Agency for Digital Italy, as notifying authority under Law 132/2025 — and of any serious incident concerning the System, and cooperate in good faith for the purposes of their respective obligations.
1. This Addendum is effective for the entire duration of the Agreement and, in the event of conflict with it, prevails as regards matters governed by the AI Act only. The processing of personal data connected with the Service, including the use of data for model training purposes, remains governed by the data processing agreement in force between the Parties.
2. This Addendum is an appendix to and an integral part of the Agreement and is concluded in writing by electronic acceptance of the Agreement on subscribing to the Service, or by express acceptance sent by certified electronic mail. For anything not governed here, the provisions of the Agreement apply, including those on governing law and jurisdiction in article 13 of the General Terms and Conditions.
For any communication concerning this Addendum, the Provider can be reached at [email protected].