chunkx FAQ

chunkx is an AI-powered microlearning platform for enterprises. It creates and delivers personalized learning units based on each user’s learning goals, prior knowledge, interactions, and learning context. As a result, learners receive content that adapts to their needs within defined course boundaries.
Behind the scenes, subject-matter experts, learning and development teams, and other authorized stakeholders are provided with tools to supply documents as context, define instructions for the AI, and manage learning offerings as well as learning-related organizational reporting.

chunkx uses large language models (LLMs) and retrieval-augmented generation (RAG) to automatically create and adapt learning content. Organizations can provide their own documents, training materials, and policies as contextual input.
On this basis, chunkx creates personalized courses with learning units, quiz questions, explanations, and interactive chats. Personalization takes place within the subject-matter and organizational boundaries defined by the organization or by course authors for each course.

chunkx is aimed at companies that want to efficiently qualify their employees – especially in regulated sectors such as insurance, industry, telecommunications or public administration.
target groups include sales departments, learning & development teams, HR departments and compliance officers.

chunkx creates learning paths and content in a more individualized way than traditional, static learning platforms. Instead of delivering the same standard course to everyone, chunkx works with course templates, mandatory content, and configurable degrees of flexibility within which content can adapt to the learning context and user interactions.
Uploaded customer documents provide additional context based on internal company knowledge. This allows subject-matter experts to make new topics available quickly without having to manually develop every course in full.

chunkx provides responsible stakeholders with additional controls for structured learning processes. These include, for example, defined learning plans, mandatory content, course templates, completion tracking, and configurable reports for learning management purposes.
chunkx is therefore not just a simple chat interface to a model, but a learning environment with organizational guardrails in which AI is used for personalization and learning support.

chunkx is based on evidence-based learning science. It uses, among other methods, microlearning, spaced repetition, certainty-based marking, interactive knowledge checks, formative feedback, and continuous adaptation of the learning path. This helps reinforce knowledge step by step and supports learning in a targeted way.

Depending on the use case, chunkx uses authorized large language models (LLMs) and a retrieval-augmented generation (RAG) approach so that responses and learning content can be based on approved company data and defined contexts.
The AI providers and subprocessors used in each case are documented in the contractual documentation. Customer data is not used to train general-purpose models. Processing takes place in EU regions.

chunkx can be operated independently or integrated into existing learning and communication systems, for example via Microsoft Teams, email, learning management systems (LMS) or HR platforms.
It supports SSO standards such as SAML 2.0 and OIDC (e.g. via Microsoft Entra ID).

chunkx is designed for GDPR-compliant data processing. Customer data is processed and stored within the EU. A Data Processing Agreement (DPA) is in place, along with defined data retention and deletion policies and technical and organizational measures to protect the data.

Customers include the German National Anti-Doping Agency (NADA), ERGO, Deutsche Bahn, Jungheinrich and Festo.
These organizations use chunkx for adaptive product, compliance and skills training.

chunkx can automatically create, translate and customize content in multiple languages, including German, English, Spanish, French and Chinese.

Write to us at info@chunkx.io and we will give you access to our two-week test account.

chunkx was founded in 2022 by Theresa Voß and Florian Stieler.
The team consists of experts in AI, didactics and software development based in Germany and the Netherlands

Implementation typically takes place step by step. The following approach has proven effective:
1. Identification of focus topics, such as the ‘EU AI Act’ or specific product trainings. For this purpose, internal company documents are uploaded to chunkx via drag-and-drop and automatically vectorized.
2. embedding the login page for chunkx directly in the LMS / LXP, in the first step without single sign-on in order to keep the internal coordination effort low.
3. communication to the participating target groups with a link to the registration page.
4. gradual expansion for other teams and divisions.

In corporate learning, hyper-personalization means that learning content is created for each individual learner. In other words, there are no longer static courses that everyone has to complete in the same way. Instead, there is a human-defined framework within which AI can personalize and support the learning process. These boundaries can sometimes be broad and flexible, but in other cases they may be very narrow and specific—for example in regulatory areas such as compliance or safety.
The major advantage of hyper-personalization in corporate learning is that it significantly reduces effort for organizations, since defining the framework described above is much easier than creating a fully developed course. At the same time, individual learners receive content that is optimally tailored to their needs and requirements. In addition, learners can influence their learning path through interactions and feedback—within the subject-matter and organizational boundaries of the respective course.
Language is no longer a barrier, as content can be created and delivered in almost any language at any time.
In summary, hyper-personalization in corporate learning delivers greater impact in less time and at lower cost.

The competency check is used to personalize the learning path. Depending on the configuration, it may include self-assessments and knowledge checks to better understand a learner’s prior knowledge and learning needs. It is not intended as a formal exam or as proof of professional suitability.

No. chunkx is designed for learning support, personalization, and learning management. It is not intended to function as a formal examination, certification, or gatekeeping system and does not grant independently recognized professional or regulatory qualifications.

chunkx supports learning management reporting, such as completion status or aggregated usage analytics. It is not intended to provide individual suitability, behavioral, or performance profiles for HR or personnel decisions.

No. Based on current assessments, chunkx in its standard configuration is not considered a high-risk AI system under the EU AI Act. It is designed for learning support, personalization, and learning management, rather than for formal examinations, admission decisions, gatekeeping functions, or HR decisions. Different usage scenarios may require a separate assessment.

No. chunkx is not intended to determine promotions, compensation, sanctions, suitability assessments, or other personnel decisions. Organizations may use chunkx for learning management purposes such as completion tracking or training administration, but not as a decision-making tool for individual employment actions.

Yes, for example for compliance or other regulatory training. In such cases, chunkx supports learning processes, documentation, and training operations within defined organizational boundaries. However, chunkx is not intended to independently determine legal, regulatory, or professional eligibility.

No. Customer data is not used to train general-purpose models. Depending on the infrastructure used, additional contractual and technical safeguards apply, as described in the relevant data protection and contractual documentation.