About

The TIB AIssistant is a platform that provides AI support throughout the entire research life cycle. You can use the various assistants sequentially—from Ideation to Review—or access each assistant individually.

When working with an assistant, you can import and export data using the assets panel on the right side of the screen. Assistants share data through these assets but do not share their context windows. Each assistant specifies a set of available tools, which you can view below the input field. It is also possible to modify the set of tools accessible to an assistant.

The TIB AIssistant is an early prototype of our vision for a research assistant that supports researchers while keeping them fully in control of their work. With this human-machine collaboration approach, we aim to provide tools that are genuinely useful throughout the research life cycle and to go beyond a simple chat interface. Future improvements will focus on community features, enabling users to create custom assistants directly through the interface.

Cite the TIB AIssistant

Please cite the following work when referring to the TIB AIssistant vision:

Auer S, Oelen A, Jaradeh MY, Khalid M, Keya F, Gaddipati SK, D'Souza J, Schlüter L, Alasti A, Rabby G, Jiomekong A., Karras, O. (2025). Towards AI-Supported Research: a Vision of the TIB AIssistant. In 5th International Workshop on Scientific Knowledge: Representation, Discovery, and Assessment (Sci-K 2025). PDF direct link.

And cite the following work specifically for the TIB AIssistant application:

Oelen, A., & Auer, S. (2025). TIB AIssistant: a Platform for AI-Supported Research Across Research Life Cycles. Joint Proceedings of Industry, Doctoral Consortium, Posters and Demos of the 24th International Semantic Web Conference (ISWC-C 2025). PDF direct link.

Source code

The TIB AIssistant is an open-source project and is released under the MIT license. We welcome contributions from the community, such as improvements to the platform or additions of new tools and assistants. You can find the source code on GitLab. The developer documentation provides an overview of how to contribute.

Demonstration video

Have a look at the demonstration video below to see how you can get started with the AIssistant. As we are continuously improving the platform, the contents of the video might slightly differ from the current version.