Listing policy
AI2F Platform – GENCI
The AI2F project (AI Factory France) aims to establish, for the benefit of the entire research community (academic, industrial, institutional, as well as startups), a single point of access listing the resources and tools available in the field of artificial intelligence.
Coordinated by GENCI as part of its public interest missions and in line with the drive for European digital sovereignty, this scheme lists the access options, whether free or paid, offered by companies, associations, institutions and other third-party entities (hereinafter the “Listed Entities”). It also provides users with a technical support team and a catalogue of specialised training courses.
The purpose of this document is, firstly, to inform any user of the AI2F platform (hereinafter the “Platform”) of the conditions and limits under which resources and tools are listed on it (« Listing Policy ») and, secondly, to set out, clearly, transparently and in a manner accessible at all times, the eligibility criteria that any Entity wishing to be listed must meet, together with the obligations arising from listing throughout its duration (hereinafter the “Policy”).
For the purposes of this Policy, the term “Listed » or « Listing » refers to the identification and recording of resources and tools within the catalogue accessible on the Platform.
Part 1 – Disclaimer
Article 1 – Nature of Listing
GENCI, as part of the AI2F project, provides users with a catalog exclusively listing the resources, tools, services, and access offered by entities that have expressly requested listing with the AI2F team and that have met, at the time of their listing, the eligibility criteria defined in Part 2 of this Policy.
The listing of a resource, tool, or entity on the Platform does not constitute:
– A recommendation, approval, technical or scientific validation, or a guarantee of quality, performance, security, or compliance of the resource or tool in question, issued by GENCI, AI2F, or any institutional funder or partner of the project ;
– A partnership, endorsement, sponsorship, or any other commercial, capital, or legal link between GENCI/AI2F and the Listed Entity ;
– A contractual commitment by GENCI or AI2F to the end user regarding the use, access, or results obtained through the Listed resource or tool.
Article 2 – Exhaustiveness and Neutrality of the Catalog
The Platform’s catalog only presents a part of the tools and resources existing within the European AI ecosystem.
The absence of listing of a resource, tool, or entity does not prejudge its quality, relevance, or compliance: it may simply result from the absence of a listing request, a request under review, or a withdrawal at the initiative of the entity itself. GENCI ensures equal, transparent, objective, and non-discriminatory treatment of any listing request.
Article 3 – Exclusive Responsibility of Listed Entities
Each Listed Entity remains solely and entirely responsible for:
– The content, availability, security, quality, and legal and regulatory compliance of the resources and tools it offers;
– The accuracy and updating of the information it communicates for the purpose of its listing and throughout its duration;
– Compliance, throughout the listing period, with the eligibility criteria defined in this Charter and any evolution thereof;
– The contractual, commercial, or technical relationships established directly with the users of its resources, to the exclusion of any involvement by GENCI or AI2F.
The end user remains solely responsible for the choice to use a Listed resource or tool and must conduct its own evaluation of the access conditions, including contractual, technical, pricing, and security conditions, offered by the relevant Listed Entity.
Article 4 – Limitation of GENCI/AI2F Liability
GENCI and the AI2F team :
– Assume no obligation of result regarding the availability, performance, accuracy, security, or absence of bias of the cataloged resources and tools ;
– Cannot be held co-responsible for any direct or indirect damage resulting from a user’s use of a Listed resource or tool on the Platform, including in the event of a cybersecurity breach, data leak, unavailability, hallucination, bias, or any other malfunction attributable to the Listed Entity;
– Reserve the right to suspend, remove, or modify at any time, without notice in the event of an established emergency (including risk to security, personal data, or the reputation of GENCI, AI2F, or the European Union), the listing of a resource, tool, or entity, without this decision engaging their liability or entitling the entity or any user to compensation.
This article does not exclude GENCI’s liability in the event of gross negligence or willful misconduct directly and exclusively attributable to it in the performance of its cataloging mission.
Article 5 – User Reporting
Any user who, in using a Listed resource or tool, encounters a breach of reasonable cybersecurity requirements, repeated data leaks, recurrent unavailability, significant hallucinations or biases (including biases contrary to ethics and good morals), or any use noncompliant with Article 5 of Regulation (EU) 2024/1689 (“AI Act”) made possible by the Listed Resource or Tool, is invited to report it without delay to the AI2F team via the procedure
available on the Platform.
Any founded report may lead to a re-examination, a precautionary suspension, or a withdrawal of the listing of the resource, tool, or entity concerned, under the conditions provided in Article 12 of this Policy.
PART 2 – Eligibility criteria for listing
Article 6 – Purpose and Scope
This Policy defines the cumulative eligibility criteria that any legal entity (company, association, public or private institution, research establishment, startup, etc.) wishing to have its resources, tools, or access (free or paid) Listed on the AI2F Platform must satisfy, as well as the applicable
obligations during the entire listing period.
It applies to any entity that has submitted a listing request to the AI2F team, as well as to any entity already Listed at the time this version or any subsequent version of the Policy comes into force.
Article 7 – Duration and Revision of the Policy
This Policy is concluded for the duration of the listing of the entity concerned on the Platform. It is consultable at any time, in its current version, on the Platform and with the AI2F team. The listing period is understood as the period of maintenance within the Platform’s catalog, from the integration of the Entity’s Resource or Tool into said catalog until the most recent event between : exclusion from the Platform and the shutdown of the Platform.
GENCI reserves the right to modify, supplement, or update the eligibility criteria defined below at any time, in particular to take into account the evolution of applicable regulations (in particular Regulation (EU) 2024/1689 on artificial intelligence, Regulation (EU) 2016/679 GDPR, or any subsequent European or national text), cybersecurity requirements, or the strategic orientations of the AI2F project.
Any substantial modification of the eligibility criteria is notified by the AI2F team to each Listed Entity by any means leaving a written trace (email, notification via the Platform). From this notification, the Entity has a period of thirty (30) calendar days or any longer period expressly indicated in the notification, where applicable depending on the nature or complexity of the compliance required, to comply with the new requirements and justify this to the AI2F team. In the absence of compliance within the given period, GENCI will proceed with the withdrawal of the listing of the entity concerned, its resources, and its tools, without this withdrawal entitling it to compensation, in accordance with Article 12 of this Policy.
Article 8 – General Eligibility Criteria
To be eligible for listing, the candidate entity must cumulatively satisfy the following criteria:
8.1 Relevance
– Offer tools, resources, or training in the field of artificial intelligence.
8.2 Identification and Legal Capacity
– Have a verified legal existence (registration, articles of association, or equivalent depending on the nature of the entity) ;
– Have the legal capacity and necessary authorizations to offer the resource, tool, or access that is the subject of the listing request;
– Designate an identified point of contact, reachable by the AI2F team for the purpose of managing the listing and any potential reporting.
8.3 Protection of Personal Data
– Demonstrate compliance with Regulation (EU) 2016/679 of April 27, 2016 (GDPR) and, where applicable, with the French Data Protection Act and any applicable national regulations on the protection of personal data;
– Guarantee the existence of appropriate legal bases for any processing of personal data carried out within the framework of the resource or tool offered, as well as adequate information for the data subjects and the effective exercise of their rights;
– Not carry out transfers of data outside the European Union;
– Host and store processed data in european based datacenters ;
– Implement reasonable and proportionate technical and organizational measures to secure the data processed and notify the AI2F team without delay of any data breach likely to affect Platform users.
8.4 Ethics and Compliance – Prohibited Uses under the AI Act
In accordance with Article 5 of Regulation (EU) 2024/1689 (“AI Act”), it is strictly prohibited for any Listed Entity to develop, acquire, deploy, market, or make accessible via the Platform an artificial intelligence system that:
– Uses subliminal, manipulative, or deceptive techniques affecting the behavior of individuals in a way that significantly impairs their ability to make an informed decision, causing or likely to cause them harm;
– Exploits vulnerabilities related to age, disability, or the social or economic situation of persons or groups of persons;
– Carries out social scoring of natural persons based on their social behavior or personal characteristics, leading to harmful or unfavorable treatment;
– Provides “real-time” remote biometric identification in publicly accessible spaces for law enforcement purposes, except in cases and conditions strictly regulated by applicable regulations;
– Infers the emotions of a natural person in the workplace or in educational institutions, except for medical or security exceptions provided by regulations;
– Creates or expands facial recognition databases through untargeted scraping of facial images from the internet or video surveillance systems.
– carries out risk assessments of natural persons in order to assess or predict the risk of a natural person committing a criminal offence, based solely on the profiling of that person or on the assessment of their personality traits and characteristics, except in the cases provided for by applicable regulations;
– carries out biometric categorisation of natural persons in order to deduce or infer their race or ethnic origin, political opinions, trade union membership, religious or philosophical beliefs, sex life or sexual orientation, except for the exceptions provided for by applicable regulations;
– generates or manipulates sexual or intimate content depicting an identifiable person without their consent, or child sexual abuse material, or is made available without reasonable and adequate safeguards where such outputs are reasonably foreseeable and reproducible.
More generally, compliance with the provisions of the AI Act and any subsequent European directive on systems, services, tools, resources, software, hardware specialized or optimized in AI, as well as their use, as soon as this directive comes into force, is a determining criterion.
8.5 Ethics and Compliance – Other Prohibited Uses
The following also constitute, under this Policy, fraudulent, illegal, or unethical uses that prevent eligibility or justify the withdrawal of listing, including but not limited to:
– Conducting analyses or processing aimed at concealing or complicating fraudulent financial transactions or money laundering operations, making their detection or traceability difficult;
– Manipulating or designing biased or deceptive AI models that could serve corrupt or fraudulent practices presented as legitimate analyses;
– Implementing social scoring practices, biometric identification in public spaces, profiling, or behavioral analysis non-compliant with regulations, or generating any content of a child pornographic nature, or infringing (i) the protection of personal data, (ii) personality rights (privacy, right to image), or (iii) respect, dignity, reputation, or moral integrity of an individual or a community of individuals;
– Automating processes intended to circumvent fraud detection or compliance systems;
– Carrying out illegal computer activities such as unauthorized cryptocurrency mining,cryptographic system cracking, or cyberattacks;
– Reselling or making available to third parties, natural or legal persons, directly orindirectly participating in the financing or support of terrorist activities;
– Processing as input data protected by intellectual property rights or confidential data without authorization;
– Generating content from confidential data or data protected by intellectual property rights for which the user does not have the necessary rights for such use;
– Conducting behavioral, qualitative, or professional or personal skills assessment of an individual, or developing personalized comments substituting human judgment, in conditions non-compliant with applicable regulations.
8.6 Cybersecurity and Quality of Service
– Implement reasonable cybersecurity measures in compliance with the AI Act, the NIS2 Directive, and the Cyber Resilience Act, proportionate to the nature of the resource or tool offered and in accordance with recognized industry standards;
– Ensure a reasonable level of availability of the resource or tool and inform the AI2F team without delay of any significant or recurrent service interruption;
– Implement reasonable mechanisms for the prevention, detection, and correction of significant biases, including unethical biases, and hallucinations, as well as transparent information for users on the known limitations of the system;
– Notify the AI2F team without delay of any security breach, data leak, critical vulnerability, or significant incident affecting the Listed resource or tool.
8.7 Preservation of the Reputation of GENCI, AI2F, and the European Union
– Not harm, through its activity, commercial practices, or public positions, the reputation, image, or interests of GENCI, the AI2F project, its institutional partners, or the European Union ;
– Not compromise, through its behavior, the proper use of public funds allocated to the financing of the AI2F project, in particular through abusive, misleading, or diverted use of listing;
– Not to be subject to any sanction whatsoever imposed by the European Union ;
– Respect, where applicable, the applicable requirements in terms of embargos, international sanctions, and restrictions related to armed conflicts, under conditions at least equivalent to those described in the ethical appendix of this Policy.
Article 9 – Listing Request Procedure
The candidate entity sends a listing request to the AI2F team via the contact form accompanied by the information and supporting documents necessary to assess compliance with the eligibility criteria defined in Article 8.
The AI2F team processes the request within a reasonable timeframe and may request any additional information necessary for its assessment.
At the end of the processing, the AI2F team notifies the candidate entity of its decision to accept, refuse, or request additional information.
Any refusal decision is justified with regard to the applicable eligibility criteria.
The listing takes effect from the publication of the resource, tool, or entity on the Platform.
PART 3 – General provisions
Article 10 – Declaration and Maintenance of Compliance
The Listed Entity declares and guarantees, at the date of its request and throughout the duration of its listing, continuous compliance with all the eligibility criteria defined in Article 8, as well as with any updated version of this Policy that may be notified to it in accordance with Article 7.
It is the responsibility of each Listed Entity to inform the AI2F team without delay of any event likely to call into question compliance with these criteria (judicial or administrative procedure, security incident, change of capital control, regulatory sanction, etc.).
Article 11 – Control and Audit
In the event of alerts from Platform users, GENCI reserves the right to request at any time from a Listed Entity justification of compliance with the eligibility criteria defined in this Policy and to carry out or have carried out, subject to reasonable prior notice, any documentary verification useful for this purpose.
The Listed Entity undertakes to cooperate in good faith with these verifications.
Article 12 – Consequences of Non-Compliance with the Policy
In the event of an established or reasonably suspected breach of an eligibility criteria or the commitments provided for in this Policy, GENCI may, depending on the seriousness of the breach and without prejudice to any other right or remedy :
– Send a request for explanation or compliance to the entity concerned, with a reasonable response period;
– Suspend, as a precautionary measure, the listing of the resource, tool, or entity concerned, in particular in the event of an established risk to user security, personal data, or the reputation of GENCI, AI2F, or the European Union;
– Permanently withdraw the listing of the resource, tool, or entity concerned, without notice in the event of a serious breach (including prohibited uses under Articles 8.4 and 8.5);
– Refuse any subsequent listing request from the same entity.
These measures are taken without prejudice to civil, criminal, or administrative proceedings that may be initiated by any competent authority or person and do not entitle the entity concerned to any compensation of any kind.
Article 13 – General Limitation of Liability
The listing of an entity on the Platform is based exclusively on the declarations and supporting documents provided by the latter.
GENCI does not carry out a systematic, exhaustive, or continuous audit of the Listed resources and tools and cannot be held liable for inaccurate, incomplete, or outdated information communicated by a Listed Entity.
In any case, GENCI’s liability cannot be engaged beyond what is strictly imposed by legal and regulatory provisions and is, in any case, limited to the consequences directly and exclusively
attributable to gross negligence or willful misconduct by GENCI in the performance of its mission as coordinator of the AI2F project.
Article 14 – Personal Data
Personal data collected by GENCI in the context of managing listing requests (identity of the point of contact and professional contact details) is processed in accordance with Regulation (EU) 2016/679 and the privacy and personal data processing policy, accessible on the Platform. This data is retained for the duration necessary for the management of the listing and, where applicable, for the justification of decisions taken under this Policy.
Article 15 – Intellectual Property
For the purposes of this clause, Intellectual Property Rights (IPR) refer to all intellectual property rights, registered or unregistered, relating to or arising from: (i) all copyrights, related rights, all computer programs (including object code and source code, as well as related documentation), databases including sui generis rights, (ii) all domain names, trademarks, trademark applications, designs, models, logos, trade names, commercial names, utility models, integrated circuits; (iii) all national or international patents or patent applications, as well as all divisions, continuations, renewals, extensions, and continuations-in-part of these patents; and (iv) all similar or equivalent intellectual property rights worldwide.
All IPR relating to the solution, source codes, databases, Tools, Resources, and training made available by the Entity remain the full and exclusive property of the Entity. GENCI does not acquire, under any title, any ownership right, license of use, or exploitation right over these IPR.
Notwithstanding the above, the Entity grants GENCI a free, non-exclusive license for the duration of the listing defined in Article 7 and for the entire world, granting the right to reference the Resource, Tool, training, or any relevant element made available by the Entity for its integration into the Platform’s catalog.
The listing keywords will be provided by the Entity. In this regard, the Entity undertakes to ensure that these keywords are not associated with third-party trademarks or IPR.
The Entity guarantees to GENCI that any Listed domain name or content of its website Listed is free of any third-party IPR and remains solely responsible for any typosquatting, cybersquatting, counterfeiting, parasitism, or unfair competition resulting from its domain name or website content providing access to the Resource, Tool, or Listed training.
Article 16 – Applicable Law and Competent Jurisdiction
This Policy is governed by French law.
Any dispute relating to the formation, interpretation, performance, or termination of this Policy that could not be resolved amicably between the parties within a reasonable timeframe falls within the exclusive jurisdiction of the competent courts of Paris, subject to public policy rules, in particular those relating to the jurisdiction of administrative courts when the nature of the dispute requires it.
Article 17 – Miscellaneous Provisions
If any provision of this Policy is deemed null or inapplicable by a competent court, the other provisions remain fully applicable. The fact that GENCI does not rely, at any time, on a breach of this Policy cannot be interpreted as a waiver of the right to rely on it later.