During the workshop, we work with use cases that municipalities will recognise. We don't present them as "success stories" — for some we are deliberately cautious — but as concrete situations to apply legal, technical and ethical analysis to. That is more instructive than an abstract discussion about "AI possibilities": participants discover for themselves where the line falls between a useful assistant and a high-risk system.
A first classic is WMO application triage. A municipality wants to shorten processing times by having AI assist with incoming applications: completeness checks, an initial indication of the allocation category, and automatic requests for missing information. The question is not whether this is technically possible — it is — but under what conditions it is permissible. A suggestion to the case handler is fundamentally different from an automated preliminary decision. We go through what that distinction means for the DPIA, for the Algorithm Register and for the objection procedure.
A second is objection classification. AI supports the sorting and summarising of incoming objections so that the legal department can respond to the substance more quickly. Here the bias question comes straight to the table: a classification that weighs postcode, surname or writing style can unknowingly push entire groups of residents to the back of the queue. We cover which evaluation steps are mandatory before such a system may go into production — and how you keep monitoring it afterwards.
A third is permit pre-screening. A spatial planning department has AI carry out a first check on a permit application: complete documentation, an indicative test against the zoning plan, flagging of risk areas. There are opportunities here, but also pitfalls — an incorrect pre-screening conclusion can wrongly send an application down the wrong track. We discuss the architecture that makes such a system auditable, including integrations with BAG, DSO and the archive.
A fourth is summarising council documents. A clerk's office or executive secretariat uses AI to draft summaries of council documents, mayor and aldermen letters or implementation documents. This is a relatively low-risk application — no decision is made about a resident — but it does come with points of attention around accuracy, hallucination and political bias. We practise with prompts that make the output more usable and easier to verify.
A fifth, and the most critical, is a civil affairs chatbot. Many municipalities are experimenting with them, but under the AI Act a chatbot that answers questions about residents' rights and obligations quickly becomes a high-risk system. We discuss the conditions under which such a chatbot can be justified, what a safe scope looks like, and what alternatives exist — for example, an internal assistant for customer contact centre staff instead of AI that residents interact with directly.
Sixth, we cover generating customer letters and decisions: AI as an assistant for staff who regularly have to draft letters and standard replies. Combined with human review and checked for plain language, this generally saves time without major legal risks. Do pay attention to archiving and traceability of AI contributions.
These use cases serve as analysis material. We go through each case in terms of AI Act classification, BIO requirements, Algorithm Register implications and practical feasibility. By the end, your team has a shared view of what it makes sense for your municipality to investigate first — and what it does not (yet).