On 10 December 2025, Visions organised, in collaboration with Prometheus-X, a webinar dedicated to a very concrete question for universities, training organisations and private higher education institutions. How can we leverage AI and data to personalise learning paths, while also connecting skills, training and employment?
This event brought together stakeholders from education, EdTech and AI around a shared belief. We cannot industrialise personalisation, nor demonstrate its impact, without a solid foundation to make data accessible, reliable and reusable. This requires an interoperable, sovereign and collaborative infrastructure.
In this context, this article presents what happened during the event: feedback from the field, use cases, demos and key takeaways.
A reality on the ground: AI needs actionable, interconnected data
In the introduction, the speakers reminded us of a simple observation: in education as elsewhere, data often remains fragmented (LMS, SIS, EdTech tools, reference frameworks, labour market/skills data). It therefore remains difficult to mobilise and govern.
To accelerate AI projects and make personalisation truly operational for training institutions, it is necessary to:
- connect existing sources (without creating yet another silo),
- harmonise and standardise data to make it reusable,
- secure exchanges (rights, consent, traceability),
- industrialise use cases that can be replicated across organisations.
It is within this framework that the Learning Data Factory and VisionsTrust fit in, together with interoperable data spaces. Taken together, they make it possible to move from a “proof of concept” logic to AI services that can be activated at scale.
3 use cases that show the value of interconnection
The event showcased this approach through real-world use cases from various sectors, demonstrating the applicability of a European technological federation in very different contexts.
Headai: more personalised, more useful and more “market-oriented” guidance
Headai presented an AI-powered guidance assistant that helps better support learners: identifying skill needs, recommending the right training (beyond the institution’s own catalogue), and linking these choices to job opportunities.
This use case interconnects LAB University of Applied Sciences, Headai, Inokufu and Edunao within a data space operated via VisionsTrust. LAB students voluntarily feed Headai’s guidance chatbot with their skills and learning history. Headai’s AI analyses this data and enriches it using training catalogues from Inokufu and Edunao, then compares these profiles with those of learners worldwide. It then generates training and skills development recommendations. The chatbot delivers these recommendations as concrete guidance and directs each student towards targeted learning pathways and job opportunities aligned with their goals.
Thus, for an institution, this enables:
- more engaging and more individualised guidance;
- better showcasing of the training offer (and of learning paths);
- a more direct link between learning and employability.
Le Cnam: VR that becomes trackable (and measurable)
Le Cnam shared feedback on immersive VR training. Le Cnam has shared feedback on its immersive VR training. This use case interconnects Le Cnam, Mimbus and Ubicast within a data space operated via VisionsTrust. Concretely, Mimbus’ VR simulator generates learning traces (xAPI) and replays, which Ubicast enriches, splits into chapters and annotates. The enhanced videos are then sent back to Le Cnam’s LMS, where teachers and students can analyse, comment on and use them to improve performance in VR.
Here, the value is not only about pedagogical innovation: it is also about the ability to track, analyse and improve learning, thanks to concrete elements (progress dashboard, chaptered video, follow-up and support perspectives).
Consequently, for an institution, this enables:
- immersive training that becomes measurable and therefore improvable;
- better support for students (feedback, follow-up);
- VR that can be integrated into a global learning environment rather than remaining a “standalone tool”.
Scheer imc: an “augmented” LMS to personalise and better steer learning
Scheer imc illustrated how an LMS can gain value when it relies on intelligent data use: more relevant learning paths, better visibility for teams, and more effective skills management.
This use case interconnects Scheer imc, MindMatcher, Schülerkarriere, Inokufu, Headai and Rejustify within a data space operated via VisionsTrust. Concretely, the skills and training data from Scheer imc’s LMS are enriched with MindMatcher’s job frameworks, Schülerkarriere’s labour market insights, and Inokufu’s analysis of learning content. On this basis, Headai’s AI models generate personalised learning path recommendations, which Rejustify aggregates, aligns and reintegrates into Scheer imc’s LMS.
As a result, for an institution, this enables:
- learning paths that are better aligned with real skills needs;
- better use of existing data, without multiplying isolated tools;
- a clearer and more actionable view of talent, gaps and development priorities.
The key: building an actionable education hub (without starting from scratch)
A key moment of the webinar focused on the very concrete approach underlying the three use cases presented above: building an education hub that combines a Learning Data Factory (Inokufu) and a data space (VisionsTrust), then quickly activating data and services via VisionsTrust.
What this hub concretely changes for an institution
With an education hub, you can:
- reduce the implementation time of data/AI projects (fewer “custom” integrations);
- increase the reliability of the data used for guidance, monitoring and personalisation;
- add new services more quickly (recommendations, dashboards, AI assistants, etc.) without starting from scratch;
- keep control over who accesses what, under which conditions, and with what level of traceability.
How it works (in simple terms)
Learning Data Factory (Inokufu): this “plug-and-play” toolbox connects to your environments (Moodle, Totara, etc.). It collects traces, anonymises and stores them in the right format. It then makes this data exploitable (dashboards, analysis, personalisation). Finally, it enables sharing results via the data space, with consent.
The use cases presented include:
- comprehensive dashboards (even outside the LMS),
- multi-source cross-analysis (jobs, LMS, Matomo, etc.),
- personalisation within the LMS,
- GDPR indicators,
- the use of LLMs without reusing learners’ data.
It also allows sharing results without disclosing students’ identities.
VisionsTrust: a technology that facilitates the connection of your tools, manages access to and use of data (rights, consents), makes offers visible via a catalogue, and enables the activation of concrete services (AI assistant / chatbot, matching & recommendations, dashboards). A live demo showed how these services can be launched from your data in just a few minutes.
Taken together, these building blocks make it possible to build an education hub to personalise learning paths, strengthen employability and accelerate the deployment of AI services at scale, without having to start from scratch each time.
In conclusion, this webinar highlighted a clear dynamic: institutions and their partners are converging towards the same priority — making learning data interoperable, governed and actionable. In this way, they can truly personalise learning paths and better connect skills, training and employment. The feedback and live demonstration confirmed it: this approach is no longer theoretical. It is already being adopted and is ready to be deployed.
If you were not able to attend the live event, the replay is available below:
If you would like to set up an education hub to connect your tools, secure and harmonise your data, and activate AI services to personalise learning paths and strengthen employability, write to us at matthias@visionspol.eu to discuss it.