AI content creation for higher education, with approval by your own institution
From module handbooks and existing material, drafts emerge for the website, leaflets, slide decks, audio and subtitles. In your own voice, with your own terminology, and published only after approval.
Where content work actually piles up
Rarely in the big campaign. Almost always in the many small texts that nobody has time for and that are still read every day.
Programme descriptions
From the module handbook and regulations a draft for the website emerges, in your own voice and with your own terminology. That version goes into approval, not straight online.
Leaflets and instructions
Re-registration, exam registration, semester abroad. The topics that generate many emails and little written material.
Slide decks for teaching
From an existing script a slide deck emerges in your visual identity, which teaching staff then adapt.
Copy for communications
News items, announcements and posts for social networks, each in the length and tone the channel needs.
Images without a stock agency
Abstract motifs and diagrams for pages and slides, when no suitable photograph exists and none should be taken.
Audio and subtitles
Texts are read aloud, recordings transcribed and subtitled. Both feed directly into accessibility.
Plain language and translation
From an approved text a more easily understandable or English version emerges. Approval stays with the responsible unit.
Consistent language across faculties
All drafts follow the same stored word list and tone, even when they are produced in six different places.
From source to approved version
Four steps, and the fourth is the one that keeps responsibility inside your institution.
Store the guidelines
Form of address, tone, approved word list and design specifications are set up once and then apply to every draft.
Choose a source
An existing document, a module handbook or a short brief. Little usable comes out of nothing.
Generate the draft
The result is a version for review, clearly marked as a draft and published nowhere automatically.
Review and approve
Someone from your institution reads it and approves it. Only then does the published version exist.
The principles we do not depart from
Generated content becomes a problem the moment nobody can say where it came from and who released it.
What the module produces
Text
Descriptions, leaflets, news items and replies, each in several lengths from the same approved basis.
Image
Abstract motifs and diagrams in your visual identity. No invented photographs of people or buildings.
Audio
Reading texts aloud and transcribing recordings, for accessibility and for following up events.
Video
Subtitles, chapter markers and summaries for existing material. The emphasis is on processing, not generation.
What HiWi+ Social Campus deliberately does not do
A content tool without a clear boundary produces material that nobody wants to answer for afterwards.
Karlsruhe University of Applied Sciences
HiWi+ is in use at Karlsruhe University of Applied Sciences and was developed together with the departments there, along real workflows rather than at a desk.
“Our main goal was to reduce administrative processes and workload within the faculty while at the same time offering students a better service, for example with the faculty chatbot they can ask questions about their studies around the clock. The HiWi+ modules were developed for our specific needs at the faculty, but they are certainly just as usable for many other educational institutions.”
What institutions ask before rolling it out
No. Every draft passes approval by a person from your institution. There is deliberately no path by which a generated text or image is published directly. That is how the module is built, not a setting.
Form of address, tone and an approved word list are stored once and apply to every draft. That is what distinguishes the module from an arbitrary text generator. The first setup happens together with your communications team.
No. Only abstract motifs and diagrams are produced. We deliberately do not generate photographs of real people or buildings, because such images are misleading and quickly become a legal problem.
Yes. Every draft is marked as generated, and the provenance stays visible: which source it came from and who approved it. That also covers the transparency obligations of the EU AI Act.
Yes, and that is one of the strongest arguments for the module. Subtitles, alternative texts, audio versions and more easily understandable versions are produced in the same step instead of as follow-up work that gets left undone.
It is processed at runtime and not fed into a model. On request, processing runs in your own data centre or entirely locally via HiWi+ Local AI.
Try it on one of your own texts
We set up your form of address and word list and produce a draft from one of your existing documents. You then see on your own material whether it sounds like your institution.