How AI Participates on MDR Academy

This page constitutes MDR Academy's transparency disclosure under Article 50 of Regulation (EU) 2024/1689 (the EU AI Act), effective from August 2025. It covers all three modes in which AI participates on this platform.

Artificial intelligence plays a real and transparent role on this platform. We use it in three distinct ways, each with different capabilities and a different scope of responsibility. This page explains all three so you know exactly what you are working with.

1. Application Chat

The in-application AI chat is available on request. To request access, contact us at mrdacademy.eu@gmail.com.

AI system used: Google Gemini API — a General Purpose AI (GPAI) model provided by Google LLC. MDR Academy integrates and deploys this model as a deployer within the meaning of the EU AI Act. Obligations that apply to the model itself — including GPAI transparency and training data obligations under Title VIII of the AI Act — are the responsibility of Google as the model provider, not MDR Academy.

Article 50(1) notification: In accordance with Article 50(1) of the EU AI Act, the chat interface displays a clear notification at the start of each session informing you that you are interacting with an AI system. This notification is shown before you submit your first message.

What the chat can do:

When you are reading a specific resource on MDR Academy, the AI can help you make sense of it — clarifying terminology, explaining the context behind a requirement, or helping you identify which knowledge categories are most relevant to your situation.

The AI can also draw on general knowledge about EU MDR 2017/745 — combining what it knows about the regulatory landscape with the content available on this platform. This means you can ask broader questions about the MDR process and receive answers informed both by our curated content and by the AI's wider understanding of the regulation.

What the chat cannot do:

The AI chat does not provide legal advice, does not review your technical documentation, and does not give binding regulatory guidance. Its answers should be treated as a starting point for understanding — not as a substitute for qualified professional advice. The AI can and does make errors, and all outputs should be verified against official sources where the matter is important.

Human oversight: If you require a human response to a question, contact Karel Boháč directly at mrdacademy.eu@gmail.com. No AI output on this platform is intended to replace human review of regulatory matters.

Training data: MDR Academy uses the Google Gemini API — not a consumer-facing Google product. Under Google's standard API terms, query data submitted via the API is not used to train Google's AI models. Your chat queries are processed to generate your response and are not retained by MDR Academy beyond your current session. For the definitive position, refer to Google's API data processing terms at https://business.safety.google/privacy/.

Ownership of AI-generated outputs: MDR Academy makes no intellectual property claim over content generated by the AI chat in response to your queries. What the AI produces during your session is yours to use. You are responsible for how you use, apply, or act on those outputs — including verifying their accuracy and ensuring they are appropriate for your situation.

2. MCP Server Access

MDR Academy exposes its knowledge base via an MCP (Model Context Protocol) server. This allows external AI agents and agentic clients — such as those built by your own team or third-party tools — to connect directly to MDR Academy's resources and query them programmatically. MCP access is available on request — contact us at mrdacademy.eu@gmail.com to discuss.

Responsibility allocation: When AI is accessed through the MCP server, responsibility for how that AI behaves, what it does with the retrieved content, and what outputs it produces lies with the operator of the agentic client — not with MDR Academy. MDR Academy provides the knowledge resource; the agentic layer is outside our control.

EU AI Act implications for MCP users: If you integrate MDR Academy content via the MCP server into your own AI system or agentic workflow, you may be acting as a deployer or provider of an AI system in your own right under Article 2(1) of the EU AI Act. This may bring its own transparency, oversight, and documentation obligations under the Act. MDR Academy's AI Act compliance extends only to the operation of this platform. Compliance for any system you build or operate using MCP-retrieved content is your sole responsibility.

If you are building an integration using the MDR Academy MCP server, please ensure your implementation handles AI outputs responsibly and does not present retrieved content as formal regulatory guidance.

3. AI-Assisted Content Authoring

The knowledge content published on MDR Academy — articles, guides, and knowledge base entries — is authored with the assistance of AI. The process works as follows: the knowledge and experience comes from people who have been through the MDR process; AI helps structure, articulate, and maintain that content in clear, readable form.

Article 50(3) disclosure: In accordance with Article 50(3) of the EU AI Act, MDR Academy discloses that knowledge base content on this platform is AI-generated text published to inform users on regulatory matters. This page constitutes the platform-wide disclosure for that content. All published articles and resources on MDR Academy should be understood as AI-assisted unless otherwise stated.

All AI-assisted content is reviewed by a human before publication. However, as with any content on this platform, it reflects experience and careful study rather than formal regulatory authority. We flag uncertainty where it exists and recommend verifying critical details with qualified professionals.

4. AI Act Compliance Framework

MDR Academy's role: MDR Academy is a deployer of AI systems within the meaning of Article 3(4) of the EU AI Act. The AI systems used on this platform are provided by third parties. MDR Academy integrates and deploys them as part of the service but does not develop or train the underlying models. Provider obligations applicable to the underlying models fall on their respective developers.

Classification: The AI chat on MDR Academy is a limited-risk AI system under the EU AI Act. It is an AI system that interacts directly with natural persons (Art. 50(1) applies), but it does not fall within the high-risk categories listed in Annex III of the Act. Specifically, it does not determine access to educational, employment, or essential services; it does not make autonomous decisions affecting users' rights or significant interests; and it is informational in nature — all regulatory and business decisions remain with the human user.

Prohibited practices: MDR Academy's use of AI does not involve any of the prohibited practices listed in Article 5 of the EU AI Act. The AI systems deployed on this platform do not use subliminal techniques, exploit vulnerabilities of any group, perform social scoring, engage in real-time biometric identification, or manipulate users against their own interests.

GPAI model context: Google Gemini, the model powering the MDR Academy chat, is a General Purpose AI model under Title VIII of the EU AI Act. Obligations specific to GPAI model providers — including transparency about training data, evaluation, and systemic risk assessment — are the responsibility of Google LLC as the model provider. MDR Academy has no role in those obligations.

Effective date: MDR Academy's obligations under Article 50 of the EU AI Act apply from August 2, 2025, in accordance with the Act's phased implementation timeline.

5. Limitations of AI on This Platform

Across all three modes of participation, the following limitations apply:

• AI can misinterpret regulatory text, produce outdated information, or omit relevant nuance
• AI outputs on this platform do not constitute regulatory or legal advice
• No AI output should be used as the sole basis for a regulatory submission, technical file, or compliance decision
• The regulatory landscape — including MDCG guidances and Notified Body practices — changes regularly; always verify against current official sources
• MDR Academy does not guarantee the accuracy, completeness, or currency of any AI-generated or AI-assisted content on this platform