Artificial intelligence governance and policies

The European Medicines Agency (EMA), as part of the European medicines regulatory network, establishes governance, policies and safeguards that help guide the use of artificial intelligence in medicine regulation.
HumanVeterinaryAIData on medicines

Danish is available via eTranslation, the European Commission's machine translation service.

Translate to Danish | Important information about machine translation

Artificial intelligence (AI) is key to leveraging large volumes of regulatory and health data.

The European medicines regulatory network aims to enable regulatory systems in the EU to use the capabilities of AI while managing its risks. To help ensure this, EMA has put in place an AI governance structure and is developing related policies.

AI capabilities in medicine regulation include: 

  • Personal productivity
  • Process automation
  • Better insights into data
  • Decision-making support for the benefit of public and animal health

Use of AI in medicine regulation has the potential to encourage research and innovation. It can also support regulatory decision-making for safe, effective and high-quality medicines that reach patients faster.

EMA guides its work on AI based on the Network Data Steering Group's (NDSG) workplan for 2025-2028. It also upholds the Network's strategy to 2028.

For more information on EMA's use of AI, see:

Network Data Steering Group workplan

The steering group's workplan identifies actions in four key AI-related areas:

  • Guidance, policy and product support – delivering guidance on the use of AI throughout the medicine lifecycle
  • Tools, technology and experimentation – providing frameworks for the use of AI tools and ensuring a structured and coordinated approach
  • Collaboration and change management – developing capacity for the use of AI technology, and informing and preparing regulators for the AI transformation

This workplan expands on the AI workplan set up by the former Big Data Steering Group.

EMA and the Heads of Medicines Agency (HMA) manage the activities of the Network Data Steering Group (NDSG). 

For more information, see: 

AI coordination group: strategy and risk assessment

EMA set up an AI coordination group to provide strategic direction for:

  • AI activities at EMA
  • Related EMA activities in support of the network

The group complements and aligns with other EMA fora and initiatives on AI, and with the NDSG.

AI System Risk Management Group

EMA's AI System Risk Management Group is a multidisciplinary operational and advisory subgroup of the AI coordination group.

The risk management group develops principles and processes to ensure a safe and compliant use of AI systems at EMA. This is in line with the EU AI Act.

The group provides guidance to EMA staff on when and how to perform risk assessment for AI systems they aim to fully deploy and use. EMA does not deploy AI tools without a through assessment process. 

The risk assessment is use-case specific and depends on factors such as:

  • AI system itself
  • AI model that the system uses
  • Data that the AI system uses
  • Target user

EMA prioritises several requirements when assessing AI tools:

  • Data protection, via data protection impact assessments
  • Information security
  • Ethical risk
  • Legal compliance
  • Necessity and proportionality – an AI tool needs to be assessed as genuinely necessary to achieve a legitimate purpose without going beyond requirements, while ensuring that its impact on individuals’ rights is proportionate and kept in check

AI guardrails and risk mitigation

EMA and the European medicines regulatory network are looking to use AI tools that are: 

  • Robust
  • Transparent
  • Fair
  • Aligned with public trust

They are taking necessary actions such as:

  • Providing so-called guardrails – they are real-time technical controls that run on every prompt or request to constrain what an AI system can accept, produce, or do; they aim to support responsible innovation in medicine development and regulation
  • Mitigating AI-associated risks – via governance of data, process management and informed AI tool roll-out

For more information on AI tools, see:

EMA policy framework on AI

EMA is preparing an AI policy framework to steer its full transition to AI-enabled work under human supervision by 2030.

The policy is needed to ensure the safe and ethical development and use of AI, in compliance with the EU AI Act.

EMA expects to publish this policy by end of 2026.

Page update history

An update log is available to show the date and summary of changes to this webpage. It does not include updates to linked documents or minor edits like typos or broken link fixes.

The tracking of updates begins in November 2026.

11 November 2026

Page first published, containing information transferred from the main AI page, such as on the Network Data Steering Group workplan

Share this page