Is the information received from the provider sufficient to enable informed use of an artificial intelligence system?
Reliance Check AI is a free guided self-assessment tool designed to help users of artificial intelligence systems understand whether the information received from the provider is actually useful for carrying out their tasks.
The tool is intended in particular for deployers of high-risk AI systems, in the context of the transparency requirements and responsibilities laid down in Regulation (EU) 2024/1689 (AI Act).
The underlying principle is simple: the fact that information appears in the instructions for use does not necessarily mean that it is sufficient to carry out a particular task.
How it works
Through a guided questionnaire, users can:
-
-
- Describe the AI system and its context of use.
- Select the tasks relevant to their activities.
- Assess whether the information received actually enables them to perform those tasks.
- Obtain a summary of information assessed as sufficient, incomplete, missing or still to be verified, with items excluded as not relevant identified separately and accompanied by reasons.
- Generate a draft request for clarification and additional information to send to the provider.
-
The result provides a structured overview of the information needs identified by the user and may support discussions with the provider and subsequent internal checks.
Based on your answers, it helps you understand whether the information received enables you to perform the selected tasks. It does not independently determine whether the information is sufficient.
The summary table may be used as internal documentation and as an informational aid for a fundamental rights impact assessment (Article 27 of the AI Act) and a data protection impact assessment (Article 35 of the GDPR or, where applicable, Article 27 of Directive (EU) 2016/680), when such assessments are required. Article 26(9) of the AI Act requires, where applicable, the information provided under Article 13 to be used when carrying out a DPIA. The table neither constitutes nor replaces such assessments, does not necessarily cover all their required elements and does not, in itself, demonstrate compliance with the corresponding obligations.
The tool is available in Italian and English.
The method
Reliance Check AI is based on a task-based approach to transparency: information needs are identified by starting from the activities that users must actually carry out, rather than merely checking whether certain categories of information appear in the documentation.
The aim is not simply to check what information has been provided, but to help users understand whether that information actually enables them to interpret outputs, use them appropriately and exercise the relevant human control and oversight functions.
The methodological approach is presented in the article by Valentina Remonato, Making AI Transparency Useful: From Information Needs to Fundamental Rights Protection, published on the CEN/CENELEC JTC 21 website on 13 October 2026.
Standardisation activities and European support
The methodology underlying Reliance Check AI is connected to the research and standardisation work carried out by Valentina Remonato within CEN/CLC/JTC 21, WG1/SAG and WG4, with support from the StandICT.eu 2029 programme.
StandICT.eu 2029 has received funding from the European Union’s Horizon Europe programme under Grant Agreement No. 101213612.
This acknowledgement concerns the support received for standardisation activities and does not imply that Reliance Check AI is a product funded, developed or approved by the European Union or by standardisation organisations.
Disclaimers and terms of use
Reliance Check AI is a tool supporting self-assessment, developed by Valentina Remonato with the assistance of AI tools, and does not constitute a legal compliance assessment, a certification, a technical assessment of an AI system or legal advice.
The results are generated solely on the basis of the information and assessments provided by the user. The tool does not independently review the provider’s documentation, verify the truthfulness of the information entered or determine the accuracy, safety, security or reliability of the AI system or of individual outputs.
The questions reflect information needs identified by the method, not an exhaustive list of the provider’s obligations. A reported gap does not automatically establish a breach of the law; a positive answer does not certify the system’s compliance. The tool does not determine the system’s legal classification or establish which obligations apply to the specific case.
Legal references are provided for guidance only. They do not imply that all the provisions cited apply to the specific case or are already applicable. Their applicability must be assessed in light of the characteristics of the system, its context of use and the dates laid down by the legislation in force, including transitional provisions.
The user remains responsible for the accuracy of the information entered, for verifying the results and for any decisions made on their basis. Requests generated by the tool must be reviewed and adapted to the specific case before they are sent to the provider.
Free use of the tool does not constitute acceptance of a professional engagement or establish a professional advisory relationship with the author.
The tool is made available, to the extent permitted by law, without any guarantee of completeness, accuracy or suitability for specific needs. It must not be used as the sole basis for operational, technical or legal decisions.
Data confidentiality
Questionnaire responses are processed locally in the browser. Reliance Check AI does not transmit them to the website’s servers. Persistent storage of responses and drafts is optional and is enabled only by selecting the relevant option; if it is not enabled, the responses remain in the open page and are not retained for a subsequent session.
Do not enter personal data, credentials, trade secrets or other confidential information. Local storage is not encrypted by this tool and does not isolate the data from other scripts running on the same origin or from anyone with access to the browser. Do not use the save option on shared devices.
The “Clear all” function deletes the questionnaire data saved by the plugin in the browser. It does not delete copies already exported, copied or printed, which remain the user’s responsibility to manage.
These statements concern questionnaire responses and the operation of the plugin, not the processing of technical browsing data or any third-party services on the website. For such processing, please refer to the website’s privacy notice.
Ownership and rights
© 2026 Valentina Remonato. All rights reserved, subject to applicable law.
Reliance Check AI is an independent tool. It is not an official product of CEN, CENELEC, UNI, StandICT.eu or the European institutions. The content and assessments expressed are those of the author and do not necessarily reflect the positions of those organisations, the European Union or the granting authority. Neither the European Union nor the granting authority can be held responsible for the content.
The tool and the results it generates may be used free of charge.
Reproduction, distribution, publication or modification of the software and protected original content requires the prior authorisation of the rights holder, without prejudice to uses permitted by law and any third-party rights or licences.