Reducing the workload in special education and optimizing the learning experience with a scalable AI strategy

Jeroen Hendriks, IT Manager at SSOE, noticed a growing challenge within special education: administrative burdens were increasing, leading to higher workloads for both teachers and administrators. This was taking away time and energy that could be spent on the primary goal: providing quality education. At the same time, he saw the potential of AI as a tool to make processes more efficient, reduce workload, and ultimately improve the quality of education.

However, Jeroen also realized that AI represented a significant shift for many staff members due to limited digital knowledge and their reluctance to embrace new technologies. Therefore, he sought an accessible way to introduce the organization to AI, to assess whether it could truly help reduce workload and to explore the best way to introduce AI for maximum adoption.

A logical first use case within the organization was reducing the workload within HR. The HR team frequently received administrative inquiries from staff, where they had to constantly refer to existing documentation. Despite having documentation available, it often took employees a significant amount of time to find the right information, which led them to eventually direct the question to HR. This made HR the perfect testing ground for the first AI solution within SSOE, with plans for expansion and other applications.

Bright Cape approached SSOE’s challenges with a strategic and innovative mindset, not only by developing a successful use case that met all requirements but also by working on a long-term AI vision for the organization.

To ensure that staff received quick answers to HR-related questions, Bright Cape developed an AI chatbot based on a Retrieval-Augmented Generation (RAG) system. This system enables the chatbot to answer questions based on internal HR documents. For example, employees can ask: ‘Am I entitled to leave if I’m moving?’ and receive an immediate, accurate response based on the available information.

The choice of a RAG system was intentional and strategic. It combines the benefits of retrieval-based and generative models, allowing the chatbot to not only retrieve relevant information but also generate coherent and contextual answers. Additionally, the open-source AnythingLLM framework provided extra flexibility and security. Since all data stays within the organization, there are no concerns regarding privacy and data protection. Furthermore, with this solution, SSOE can develop and customize their own chatbots. For instance, the organization recently created a chatbot for an internal event—a great example of the versatility and scalability of the solution.

However, building the chatbot was just one part of Bright Cape’s added value. Equally important was supporting SSOE in shaping their AI vision. Bright Cape helped the organization identify the key risks and requirements that AI solutions in education must meet. Through this approach, SSOE was able to implement a safe and scalable AI solution while also taking steps on their AI roadmap for the future and creating a moral compass for the use of AI within special education.

The implementation of the AI chatbot has significantly improved internal processes at SSOE and increased efficiency. By using the open-source AnythingLLM framework, the data remains secure within the organization, and staff have the freedom to build and customize their own bots.

Thanks to the chatbot, the time HR employees spend on repetitive questions has been drastically reduced, saving them dozens of hours per month. As a result, the investment was recouped within just a few months. But perhaps even more importantly, the pilot has shown that AI can indeed contribute to reducing workload in education. This has provided Jeroen and SSOE with confirmation that AI can play a valuable role, as long as it meets certain requirements and is deployed in a safe and accessible way.

With these insights, SSOE can now continue to build on broader AI adoption within the organization. By gradually testing new applications and involving staff in the possibilities, AI is being integrated into the educational process in a sustainable manner