Using Generative AI (GenAI4EU ) for Scientific Research via EOSC

The EU funding opportunity titled "Using Generative AI (GenAI4EU) for Scientific Research via the European Open Science Cloud (EOSC)" falls under the Horizon Europe program, specifically HORIZON-INFRA-2025-01-EOSC-05. This initiative aims to enhance scientific research capabilities by integrating generative AI technologies into the EOSC ecosystem. The grant is categorized as a HORIZON Research and Innovation Action and is allocated as a lump sum grant, with funding expected to range between €7.5 million and €10 million per project. The application process is a single-stage submission, with a deadline set for September 18, 2025.

Eligible applicants include universities, research institutions, small and medium-sized enterprises, and large enterprises, particularly those involved in AI and data science. The geographic eligibility encompasses EU member states and associated Horizon Europe countries. The projects funded under this call are expected to focus on demonstrating and fostering the utilization of generative AI throughout the research data lifecycle, supported by the FAIR data principles.

The grant aligns with the GenAI4EU initiative and emphasizes the operationalization of generative AI tools to improve data processing, analysis, and reporting in scientific research. It encourages collaborations among multidisciplinary teams and aims to enhance data sharing and accessibility while ensuring compliance with ethical standards.

Projects funded through this opportunity are expected to produce high-quality, machine-readable datasets suitable for generative AI applications, contribute to the creation of common platforms for secure data sharing, and engage in community outreach to enhance understanding and readiness for AI tools in scientific workflows. Additionally, the initiative emphasizes the significance of open data practices, transparency, and proper documentation to foster trust and maintain data integrity.

Overall, this funding opportunity aims to push the boundaries of scientific research capabilities by leveraging generative AI, facilitating cross-disciplinary applications, and improving the usability and availability of data across various scientific domains.

The EU funding opportunity HORIZON-INFRA-2025-01-EOSC-05, titled "Using Generative AI (GenAI4EU) for Scientific Research via EOSC," aims to demonstrate and foster the use of Generative AI in scientific research, aligning with the GenAI4EU initiative and other key EU strategies like the Apply AI strategy, throughout the research data lifecycle supported by the European Open Science Cloud (EOSC). The call falls under the Horizon Europe (HORIZON) program, specifically the Research Infrastructures 2025 call (HORIZON-INFRA-2025-01). It is a HORIZON Research and Innovation Action (HORIZON-RIA) with a HORIZON Lump Sum Grant [HORIZON-AG-LS] type of Model Grant Agreement (MGA).

The call is currently open for submission as a single-stage process, with an opening date of May 6, 2025, and a deadline of September 18, 2025, at 17:00:00 Brussels time. The total budget for this topic is 37,500,000 EUR, and the indicative number of grants to be awarded is 4, with contributions ranging from 7,500,000 to 10,000,000 EUR.

Expected outcomes of the projects funded under this call include:

EOSC making available high-quality, machine-readable scientific datasets for consumption by machine-driven Generative AI applications, benefiting science and aligning with the GenAI4EU initiative and the Apply AI strategy.

EOSC facilitating the pooling and sharing of high-value datasets from EOSC and other priority data spaces, such as public sector, health, climate, environmental, manufacturing, agriculture, energy, financial, and mobility data.

Large-scale actions supported by EOSC, including the creation of common data platforms for secure and compliant sharing and reuse of sensitive, confidential, proprietary, and personal data, as well as large-scale experimentation based on Generative AI, in line with the GenAI4EU initiative and the Apply AI strategy.

The scope of the call focuses on demonstrating and fostering the use of Generative AI for scientific research, improving productivity through activities like writing, data generation and analysis, and reporting. This aims to elevate science beyond human limitations by deploying smart algorithms, machine learning, and AI services onto the Web of FAIR Data. The call also emphasizes raising awareness and readiness for using Generative AI in scientific research through training activities.

AI-powered natural language interfaces are expected to transform how researchers interact with open science infrastructures, discover, and combine relevant data, software, and application assets. EOSC should evolve to offer unbiased and trustworthy responses, adopting FAIR practices for AI-trained models to address reproducibility and trustworthiness challenges.

Open Data and Open Research Software are considered essential for reliable, trustworthy, and transparent GenAI, ensuring datasets and algorithms are well-documented, accessible, and reproducible. This transparency fosters trust, supports ethical standards, and ensures compliance with regulations, particularly important in the field of GenAI.

Proposals should focus on the following aspects:

Enriching the EOSC federation with Generative AI tools for evaluating research data quality and ensuring trustworthiness across the European network of trusted repositories. This includes formulating protocols and policies to facilitate effortless data access, processing, and provenance updates within EOSC's repository and service network.

Supporting European research infrastructures to improve the FAIRness of their data, enabling combination with data from infrastructures in neighboring scientific domains to provide Generative AI-ready data. This involves conducting pilots to validate the effectiveness and accuracy of Generative AI-driven data quality evaluation methods, iteratively improving them based on feedback and real-world use cases, and removing potential biases inherited from the training data.

Running community engagement and support programs for implementing Generative AI in scientific workflows via EOSC, promoting training programs to facilitate the uptake and use of Generative AI for FAIRification of data and data curation, demonstrating how Generative AI can facilitate quality assessment of FAIR data, advancing the realization of machine-actionable (MA) research data and services, including AI-based systems, and proposing protocols and policies to govern automatic data workflows within the network of repositories and services.

Proposals are expected to deliver on one or more of the following:

Developing, promoting, and supporting real-life use cases for Generative AI models in scientific research domains, such as:

Augmenting datasets in scientific fields that rely on image analysis (e.g., biology, astronomy, materials science) by generating synthetic images that closely resemble real data, improving model robustness, sharing anonymized versions of sensitive data, and generalizing better to unseen scenarios.

Learning the underlying patterns of complex time-series data (e.g., sensor readings in environmental monitoring, physiological signals in healthcare) by generating data samples that match the learned distribution, detecting anomalies or deviations from normal behavior.

Accelerating materials design and discovery by predicting the properties of new materials without extensive experimental testing, generating novel material structures with desired properties based on learned relationships between material compositions and properties.

Advancing drug design and molecular modeling by generating novel molecular structures with desired pharmacological properties, exploring vast chemical spaces, predicting interactions between molecules and biological targets, and optimizing drug candidates for efficacy and safety.

Simulating complex systems and phenomena in various scientific domains (e.g., physics, chemistry, ecology) by capturing the underlying dynamics and interactions of the system, generating realistic simulations that mimic observed behavior or predict future outcomes under different conditions.

Proposers should leverage results from relevant projects like AI4EOSC, iMagine, EOSC Data Commons, RI-SCALE, and other developments within the GenAI4EU initiative and the Apply AI strategy. The Joint Research Centre (JRC) research infrastructure may be included in proposals for creating and sharing high-quality machine-readable scientific datasets.

General conditions for participation include:

Admissibility conditions related to proposal page limits and layout, as described in Annex A and Annex E of the Horizon Europe Work Programme General Annexes and Part B of the Application Form.

Eligible countries as described in Annex B of the Work Programme General Annexes, with specific provisions for non-EU/non-Associated Countries as detailed in the Horizon Europe Programme Guide.

Other eligible conditions, including restrictions for the protection of European communication networks and the potential participation of the Joint Research Centre (JRC).

Financial and operational capacity and exclusion criteria as described in Annex C of the Work Programme General Annexes.

Evaluation and award criteria, scoring, and thresholds as described in Annex D of the Work Programme General Annexes.

Submission and evaluation processes, with additions to the general award criteria considering the coordination efforts and resources with other relevant projects and the EOSC governance structure, as described in Annex F of the Work Programme General Annexes and the Online Manual.

Indicative timeline for evaluation and grant agreement as described in Annex F of the Work Programme General Annexes.

Legal and financial setup of the grants, including additional access rights for beneficiaries, such as granting royalty-free access to results and intellectual property rights to the EOSC Association and legal entities identified by the granting authority, depositing digital research data in a trusted repository federated in EOSC, and eligible costs taking the form of a lump sum.

Grants awarded under this topic will be linked through collaboration agreements to the grants from the HORIZON-INFRA-2025-01-EOSC-03 action.

Specific conditions are described in the specific topic of the Work Programme.

Application and evaluation forms, model grant agreement (MGA), and additional documents are available in the Submission System and include:

Standard application forms (HE RIA, IA, HE CSA).

Standard evaluation forms (HE RIA, IA, HE CSA).

Guidance documents such as the HE Programme Guide, Model Grant Agreements (HE MGA, HE Unit MGA, Lump Sum MGA, Operating Grants MGA, Framework Partnership Agreement FPA), call-specific instructions, detailed budget table (HE LS), information on financial support to third parties (HE), information on clinical studies (HE), and guidance on lump sums.

Additional documents include the HE Main Work Programme 2025, HE Programme Guide, HE Framework Programme 2021/695, EU Financial Regulation 2024/2509, Decision authorising the use of lump sum contributions, Rules for Legal Entity Validation, EU Grants AGA, Funding & Tenders Portal Online Manual, and Funding & Tenders Portal Terms and Conditions.

The call aims to leverage Generative AI to enhance scientific research within the EOSC framework, promoting data quality, FAIRness, and the development of innovative applications across various scientific domains. It seeks to foster collaboration, training, and the establishment of protocols and policies to govern the use of AI in scientific workflows, ultimately contributing to the advancement of science and innovation in Europe.

In summary, this funding opportunity is geared towards projects that can effectively integrate Generative AI into the scientific research process via the European Open Science Cloud. It calls for proposals that not only develop and implement AI tools but also address the critical aspects of data quality, trustworthiness, and ethical considerations. The goal is to empower researchers with advanced AI capabilities, streamline workflows, and unlock new possibilities for scientific discovery across diverse fields. The emphasis on FAIR data principles, training, and community engagement underscores the commitment to creating a sustainable and inclusive ecosystem for AI-driven research in Europe.

Eligible Applicant Types: The eligible applicant types are not explicitly defined in the provided text. However, based on the nature of Horizon Europe Research and Innovation Actions (RIA) and Coordination and Support Actions (CSA), eligible applicants typically include universities, research institutes, SMEs, large enterprises, and other organizations capable of conducting research and development activities. The Joint Research Centre (JRC) of the European Commission can also participate as a member of a consortium.

Funding Type: The primary funding mechanism is a grant, specifically a HORIZON Lump Sum Grant, under the Horizon Europe Programme. There are also Coordination and Support Actions.

Consortium Requirement: The opportunity appears to require a consortium, as the Joint Research Centre (JRC) may participate as a member of the consortium.

Beneficiary Scope (Geographic Eligibility): The geographic eligibility includes EU member states and countries associated with the Horizon Europe Framework Programme. Non-EU/non-Associated Countries may also be eligible if they have made specific provisions for funding their participants in Horizon Europe projects, as detailed in the Horizon Europe Programme Guide.

Target Sector: The program targets the following sectors: artificial intelligence (AI), specifically Generative AI, scientific research, data science, cloud computing, and research infrastructures. It also touches upon various scientific domains that can benefit from Generative AI, such as biology, astronomy, materials science, healthcare, drug design, molecular modelling, physics, chemistry, and ecology.

Mentioned Countries: The opportunity explicitly mentions the European Union (EU) and non-EU/non-Associated Countries.

Project Stage: The expected maturity of the project is focused on demonstration and fostering the use of Generative AI for scientific research, which suggests a stage of development, validation, and demonstration. The call aims to enrich the EOSC federation with Generative AI tools and support European research infrastructures in improving the FAIRness of their data.

Funding Amount: The funding range varies depending on the specific topic within the call:
HORIZON-CSA: €1,000,000 to €8,000,000
HORIZON-RIA: €3,000,000 to €15,000,000
The total budget for the call is approximately €329,000,000.

Application Type: The application type is an open call, with a single-stage submission process.

Nature of Support: Beneficiaries will receive money in the form of a lump sum grant.

Application Stages: The application process is single-stage.

Success Rates: The success rates are not explicitly mentioned in the provided text.

Co-funding Requirement: The text does not explicitly mention a co-funding requirement.

Summary: This Horizon Europe call (HORIZON-INFRA-2025-01-EOSC-05) aims to promote the use of Generative AI in scientific research by providing grants to consortia that can demonstrate and foster its application throughout the research data lifecycle within the European Open Science Cloud (EOSC). The call encourages projects that will enrich the EOSC federation with Generative AI tools, support European research infrastructures in improving the FAIRness of their data, and run community engagement and support programs for implementing Generative AI in scientific workflows. The call is part of the GenAI4EU initiative and seeks to leverage Generative AI for activities such as writing, data generation and analysis, and reporting, to improve productivity and enable science beyond the human scale. Eligible applicants include universities, research institutes, SMEs, and other organizations from EU member states and associated countries. The funding is provided as a lump sum grant, with projects expected to contribute to making high-quality, machine-readable scientific datasets available for Generative AI applications, facilitating the pooling and sharing of high-value datasets, and creating common data platforms for secure and compliant data sharing. The call also encourages the inclusion of the European Commission's Joint Research Centre (JRC) research infrastructure in project proposals. The deadline for submission is September 18, 2025.