Schulinck AI Desk

Municipality of Oss

An AI assistant that helps local authority staff with legal queries relating to the social services sector.

Last change on 17th of August 2026, at 11:14 (CET) | Publication Standard 1.0

• Publication category: Other algorithms
• Impact assessment: Field not filled in.
• Status: In use

General information

Theme

• Social Security
• Health and Healthcare

Begin date

2026-07

Contact information

[e-mail]

Link to publication website

https://oss.nl

Responsible use

Goal and impact

Objective: Schulinck AI-Desk has been developed to provide local authority professionals (such as advisers, legal administrative staff and quality assurance officers) with immediate, clear answers to legal queries. This enables staff to work more efficiently and reach the correct solution more quickly in ongoing cases. Impact: In day-to-day use, AI-Desk supports professionals in areas such as handling individual requests for assistance and ensuring legal quality internally. This results in more consistent and meticulous service provision to residents.

Considerations

AI helps professionals to find relevant information and sources more quickly, to interpret legal matters, and to learn and develop their professional expertise. The responsibility for weighing up answers and exercising human judgement remains with the professional. AI is therefore not a substitute for the professional, but a tool.

Human intervention

Schulinck AI-Desk does not make independent decisions. Decision-making always remains in the hands of a professional. Whilst Schulinck AI-Desk does provide answers to questions, it functions solely as a tool for staff members, not as an automated decision-maker. In practical terms, this means that a local authority professional uses the advice or answer provided by AI-Desk to support their own judgement. The staff member critically assesses the output, checks the sources where necessary, and applies the information to the resident’s specific situation. For example, if the algorithm takes a particular legal stance, the civil servant will still assess for themselves whether this is comprehensive and appropriate in the context of the case.

Risk management

No personal data is processed in the AI Desk.

To answer questions, only a combination of municipal and Schulinck-specific content is used, which has been checked for accuracy by Schulinck. In addition, the quality of the answers provided by the Schulinck AI Desk is constantly monitored by Schulinck and improved where necessary.

Legal basis

The AI desk is used solely for internal processes and does not make decisions without human intervention. A legal basis is therefore not required.

Elaboration on impact assessments

No personal data is processed. This solution serves as a knowledge base

Operations

Data

Field not filled in.

Technical design

Schulinck AI-Desk is based on generative AI using what is known as a retrieval-augmented generation (RAG) architecture. This means that the system combines the power of a large language model (LLM) with targeted information retrieval from its own knowledge database. Broadly speaking, this works as follows:

1) Question input

2) A user enters a question in natural language into Schulinck AI-Desk, for example: “What are the conditions for refusing a bespoke provision under the Wmo?”. The system receives this query.

3) Search for sources

4) Schulinck AI-Desk searches in various ways for sources that are well suited to the question asked. The result is a selection of relevant text excerpts, for example an article from the Youth Act and an explanatory note.

5) Answer generation using LLM

6) Schulinck AI-Desk provides an answer to the question based on the sources found. In effect, the LLM functions here as a language processor that summarises and explains the legal sources in understandable terms. The output usually consists of one or more paragraphs answering the question, including legal reasoning and references to the sources used

7) Follow-up

8) The user can ask follow-up questions within the same context. This ensures that follow-up questions take into account the information previously found

External provider

Wolters Kluwer