Skip to main content

Built with a German university.
Sources you can check.

LearnSlice is co-developed in a publicly funded research project with FH Dortmund. Here is what the first university pilot looked at, and what it found.

100%
Every answer traceable to a source
80%
Reached near-complete mastery
70%
Found the learning pace appropriate
FH Dortmund
  • Co-developed with FH Dortmund
  • Federally funded research (BMWE)
  • Hosted in Germany

What the study looked at

LearnSlice was co-developed with the IDiAL institute at FH Dortmund in a research project funded by the German Federal Ministry for Economic Affairs and Energy (BMWE, grant 16GM200302). In May 2026, 50 master's students used it to prepare for an exam in a Usability Engineering course. The evaluation used a Design Science Research approach with two student surveys and a focus group of five lecturers with 8 to 19 years of teaching experience. This is the first of three planned iterations.

Where students said exam prep is hardest

Before the pilot, students named their biggest exam-preparation challenges. The didactic design responds directly to the top ones. 60% of these students already studied with AI before the pilot, which is exactly why verifiable sources matter.

  • Understanding theoretical concepts 22%
  • Applying knowledge to practical tasks 18%
  • Staying motivated 16%
  • Time management 12%
  • Structuring their learning 10%

Established didactic methods, not just answers

LearnSlice combines curated curriculum content with proven teaching methods, configured by the research team.

Socratic dialogue

Guided questioning that helps learners reason toward an answer instead of only receiving it.

Active recall with quizzes

Short quizzes and flashcards that turn passive reading into practice and self-checking.

Curriculum-anchored answers

Retrieval grounded in your materials, with inline citations back to every source passage.

Gamification, used in context

Game elements add motivation where they fit. The experts noted these work best in specific situations, not as the sole motivator.

The next iterations add project-based learning and the integration of company-specific knowledge.

What the first pilot showed

100%

Every answer traceable to a source

In the system evaluation, every answer carried a valid citation back to its source passage, with no invented references.

80%

Reached near-complete mastery

After preparing with LearnSlice, 80% of students reached complete or near-complete understanding of the exam topic.

70%

Found the learning pace appropriate

70% of students felt the personalized learning pace matched their individual needs.

Where students said LearnSlice helped more

Asked where LearnSlice gave them more than general AI tools, students most often pointed to exam focus and structure.

  • Exam-oriented support 36%
  • Structured learning guidance 18%
  • Relevant answers 16%
  • Practical examples 14%

A curriculum-anchored, sovereign approach

  LearnSlice General AI tools
Data-processing location Hosted in Germany Vendor cloud, region varies, often outside the EU
Learner data used for training No Governed by vendor terms, varies by plan
Primary knowledge source Your curriculum and documents General model knowledge
Source attribution Inline citations to retrieved passages Not provided by default
Cost model Fixed self-hosted infrastructure Usage-based per-token pricing

What this means for your role

For leaders

A credible, compliant choice: co-developed with a German university in a federally funded project, with learner data hosted in Germany and never used to train the model.

For teachers

It answers from your curriculum, and every statement links back to the source passage. You keep control of the content, with no invented references.

For HR and L&D

Built for the skills gap in vocational training. You can feed it your own and regulated material, with a roadmap toward company-specific knowledge.

“A tailored assistant adds the most value exactly where training needs it most: specialized, regulated, and curated-source domains.”
Prof. Dr. Carsten Wolff, FH Dortmund, IDiAL institute

What we are still working on

This was a first-iteration pilot on a foundational topic, and the gains in learning speed and engagement were partial. The lecturers noted that the value of a tailored assistant is highest for specialized, regulated, and curated-source material. The next iterations focus on project-based learning and integrating company-specific knowledge. We will update this page as the research continues.

First-iteration research pilot, 50 students, one course, May 2026. LearnSlice is a partner in the funded research project. Findings describe an early study and are not a peer-reviewed publication.