AI in debt collection: where it helps and where people decide

Finance teams often hear AI described as the next big promise. Once it gets specific, the fair question follows: what does the AI actually decide, and who is responsible? In debt collection this is not a side issue, because every step towards a debtor has legal consequences.

That is why incaseof.law keeps three levels clearly apart. AI reads documents. Fixed rules run instalment plans. And people decide on dunning stages and the route to court. This separation makes the process fast and easy to follow.

Who decides what at incaseof.law?

AI reads supporting documents from the customer portal and court requests to correct a filing, and the case team checks the result. Instalment plans and payment reminders follow fixed rules that are the same for every debtor. Dunning stages, the recommendation to file a payment-order claim and every step towards court sit with the team, and nothing is filed until you approve it.

incaseof.law, as a licensed debt collection agency, is responsible for the out-of-court steps. Court steps require your approval. The graphic shows where each level comes in.

What does AI read at incaseof.law?

AI at incaseof.law reads two kinds of documents: the supporting documents you upload in the customer portal when you hand over a claim, and court requests to correct a filing. In both cases the team checks the result before it is used.

Documents at handover. Invoice, delivery note, order confirmation: AI captures the details the case needs, such as amount, dates and debtor, so nobody has to retype them. The team checks the captured data before the case moves on. Documents submitted through an interface are read by our team.

Court requests to correct a filing. If a court objects to a filing, for example because a detail is missing or unclear, it issues a request to correct it within a deadline (in Austria, a Verbesserungsauftrag). AI reads out what the court requires and by when. The team checks the result and deals with the request.

There is also an often overlooked cost in collection: switching between systems. Data from systems such as SAP or BMD has to be prepared so that the courts accept it through electronic court filing (ERV). The platform takes care of this preparation. Nothing is filed until you have approved court proceedings. For larger volumes, the same logic runs directly from your own system, as the Enterprise integration provides.

The table sums up what AI does at incaseof.law and what it does not do.

Overview: incaseof.law, as of October 2026. AI is used only to read documents.

Why does incaseof.law use AI only for reading?

Because every step in debt collection has legal consequences. A dunning stage, a claim or enforcement affects the debtor’s rights, and someone who knows the case has to stand behind it. That is why people decide at incaseof.law, and AI takes the reading work off their hands.

The speed comes from the fact that nobody has to retype documents or evaluate court letters by hand. That leaves the team more time for the decisions that matter.

Late payments weigh on the liquidity of European businesses, as Intrum’s European Payment Report 2024 shows. The most expensive mistake with unpaid invoices remains waiting: a claim loses value as it ages, and in Austria many claims for goods and services become time-barred after three years.

Does the process stay fair to debtors?

Yes. The same fixed rules apply to every debtor for instalment plans, there are no risk classes and no assessment of individuals. The team decides on dunning stages and the route to court case by case.

A debtor who wants to pay can request an instalment plan in the debtor portal and receives a reminder before each instalment is due. If an instalment is still unpaid after the grace period, the instalment terms are lost under the same fixed rules and the remaining balance falls due (Terminsverlust). If a debtor does not respond, no time is lost, because the team decides on the next dunning stage. For how the process is organised, see Digital receivables management; the handover process is explained in How debt collection works.

What is incaseof.law working on?

We are working on analyses that could help our case team decide in which order cases are handled. This is not in use today, and even then the decision on dunning stages and the route to court stays with the team.

Before anything changes in how data is processed, we first update the privacy policy, the data processing agreement and the data protection impact assessment. Only then do we change the process and this page.

This article describes how the incaseof.law platform works in general and does not replace advice on individual cases. As of October 2026. Plans described as an outlook are not in use today.