DFG requests ideas for the conception of funding measures in the area of Scientific Library Services and Information Systems (LIS)

Objective

The Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) is seeking to tap into the huge potential offered by the use of artificial intelligence (AI) in research by developing suitable funding measures in close cooperation with the research community. The aim of this call is to collect ideas for funding measures to support AI in research. The ideas can relate to any of a whole range of different aspects (technical, organisational, legal, etc.), but they should be feasible for implementation by scientific information infrastructure facilities such as data centres, libraries, computer centres, archives, museums and the like. An idea may involve enabling, facilitating or improving the use of AI, or making the use of AI more transparent. An idea can also address how AI can be used for tasks in scientific information infrastructures. This call for ideas is not concerned with promoting basic research into AI.

Context

Artificial intelligence – understood here with an emphasis on machine learning methods and on software that generates language and images – is used in research practice for various purposes and in pursuit of various objectives. There are certain requirements to ensure that such use upholds good research practice and the quality of research results. Scientific information infrastructures have the capacity to establish a range of prerequisites for the use of AI in research (e.g. provision of training data, design and establishment of instruction and training measures, organisation of access to computing resources); such infrastructures are also capable to ensure transparency, organise reuse (e.g. domain-specific models based on FAIR principles) and much more besides. 

Target group

The call for ideas aims at anyone who uses, develops and/or operates scientific information infrastructures. Ideas for funding measures can also be proposed by scientific or infrastructural networks, associations and organisations. The call for ideas is open to all career levels and research institutions in Germany – universities, non-university research institutions, universities of applied sciences and others.

Format

Ideas can be submitted in German or English on a maximum of three pages in PDF format. The submission must have a title and include the name of a contact person, along with the names of any other persons involved. There is no requirement to submit letters of support, CVs or other attachments.

Structure

First, please state the infrastructural aspect that is at the core of the funding idea – e.g. technical, organisational, competence-building or legal, but also in terms of content e.g. with regard to training data, data curation and much more. After this, describe the context in which AI can be used for research in the subject areas addressed. In addition, attempt to estimate who would benefit and to what extent. You can then provide concrete examples to illustrate your funding idea. Finally, assess the prospects of success, the difficulties to be overcome, the resources required (staff and direct project costs) and the time frame for realising your funding idea.

Evaluation

The ideas submitted will first be assessed in terms of their suitability for LIS funding and then be evaluated by the Committee on Scientific Library Services and Information Systems (AWBI). The key selection criteria are firstly whether an idea offers potential for utilisation in at least one research area, and secondly whether it enables scientific information infrastructure facilities to position themselves and make progress on areas that are relevant to the support of AI in research.

Result

After the evaluation, the contact persons will receive feedback on their funding idea. The most promising ideas will be taken as a basis for funding measures to be offered from 2025 onwards, for example in the form of calls for proposals.

Submission and deadline

The contact person is to submit the document via the elan portal by no later than 13 September 2024. Please note that this requires registration on the elan portal, which is usually confirmed by the following working day.

Further information

Statement by the Executive Committee on the influence of generative models of text and image creation on science and the humanities and on the DFG's funding activities:

• www.dfg.de/statement_executive_committee_ai

Funding area “Scientific Library Services and Information Systems (LIS)”:

• www.dfg.de/lis/en

DFG Programme Contacts