Artificial Intelligence in Higher Education: The Future is Now!

Artificial Intelligence in Higher Education: The Future is Now!

The integration of generative artificial intelligence (GenKI) and large language models (LLMs) in higher education has become increasingly important in recent years. Particular attention is being paid to the European ADMIT project, which presented the first results of a comprehensive survey on June 13, 2025. The aim of the project is to develop ethical guidelines and didactic principles for the use of GenKI and LLMs at universities in order to meaningfully integrate these technologies into teaching. The project is carried out by the FernUniversität Hagen under the direction of Prof. Dr. Claudia de Witt and Silke E. Wrede pushed forward and coordinated by the EADTU. FernUniversität Hagen reports that...

ADMIT, which stands for “generative AI and large language models in higher education”, includes eleven distance and open universities in Europe. A central aim of the project is to collect best practice examples and develop a framework for the responsible integration of LLMs and GenKI into teaching and learning contexts. This includes the creation of an open access online course that is intended to provide practical information on the opportunities and challenges of GenKI. This initiative is supported by the European Commission and the FernUniversität's internal research funding.

Surveys and analysis of the use of GenKI

The initial survey results provide insights into teachers' individual, institutional and national perspectives on GenKI and LLMs. In order to take developments and changing usage patterns into account, a new round of surveys is planned in June to record feedback and adjustments in the context of ethically sound usage. The surveys analyze how these technologies influence the educational landscape and which ethical issues need to be taken into account.

In addition, the need for a deep understanding of existing strategies to create sustainable applications of GenKI is emphasized. The FernUniversität shows a strong commitment to topics such as diversity, equality and inclusion in connection with the use of these technologies.

Technological aspects of LLMs

In terms of technology trends, various sources provide information about the functionality and applications of LLMs such as ChatGPT, GPT-4 and Llama 2. These powerful human language processing and generation models are increasingly being used in natural language processing, content creation and customer support companies, often without the need for additional fine-tuning. Fraunhofer IESE reports that...

When selecting an LLM, criteria such as technical compatibility, adaptability, costs as well as the legal and ethical implications are of central importance. The importance of sustainability considerations is also highlighted: companies should consider whether LLMs are necessary for their needs or whether smaller, less complex models are sufficient. Careful implementation also requires a structured approach that includes defining the specific tasks and identifying the relevant data.

Educational policy support and interdisciplinary approaches

The promotion of artificial intelligence in higher education also comes from the Federal Ministry of Education and Research (BMBF), which has been supporting projects aimed at the development of study programs and AI-based learning systems since the end of 2021. These initiatives not only promote the technology, but also the skills and infrastructure required for it. Education Server reports that...

Numerous AI projects in the education sector are being developed in an interdisciplinary manner to improve teaching and learning processes. Dissemination projects such as Cyber ​​Valley and the AI ​​Campus provide additional resources and materials to advance the integration of AI into higher education. The Leopoldina's lectures and discussion papers as well as the BMBF's AI action plan illustrate the discourse about the responsible development of GenKI and its role in the educational landscape.