Skip to content

Kipo Analyzer eases Lower Saxony police review of abuse images

Lower Saxony's State Audit Office reports that AI tools that pre-sort seized abuse images, used by the state police since 2020, considerably reduced investigators' workload.

  • 1 for an organization

    The auditor reports a considerable reduction in workload and clear time savings, from users' experience, without a figure.

  • Not rated a person

    The auditor states that pre-sorting lowers the psychological strain of viewing abuse images and gives no measure of it.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudied, the grade of this recordreportedclaimed

An independent research report supports the benefit.

The system’s part
Not isolated. The workload relief comes from the Interior Ministry's January 2025 evaluation of users' experience, which the auditor adopted. The report gives no figure and no comparison of review with and without the tools. It says that the police trained the tools on a dataset from a criminal case and that they need regular retraining, so they learn from data.
Not yet checked
LimitsCosts and harmsLearns from dataLater outcome
Benefit type
Reduced Task Time
System
Kipo Analyzer and Tracebook Kipo
Made by
Landeskriminalamt Niedersachsen (LKA, the state criminal police office)
Used by
Polizei Niedersachsen (LKA investigations of child and youth sexual abuse material)
Where
Lower Saxony (Niedersachsen), Germany
Stage
In routine use
Date
, when the source was published
Kind
Program
Broad benefit
Identifying Objects, Actions, and Events, Processing Information
Last revised
1 November 2026

Sources

  1. Report, by authors with no tie to the system.

    Niedersächsischer Landesrechnungshof. Jahresbericht 2026 zur Haushalts- und Wirtschaftsführung, Beitrag 25, Kinderpornografie mit Künstlicher Intelligenz wirksam bekämpfen. Presented to the Landtag on 10 June 2026. Printed p. 171 (PDF p. 176). The office takes the workload relief from the Interior Ministry's evaluation of January 2025.

    KI unterstützt die Polizei bei der Aus- und Bewertung von Bildmaterial. Dies führte zu einer erheblichen Arbeitsentlastung.
    Den Einsatz der KI-Software bewerteten die Betroffenen als arbeitserleichternd. Die KI-Lösungen führten zu einer deutlichen Zeitersparnis und Entlastung.
    Ohne regelmäßiges Nachtrainieren besteht die Gefahr, dass KI-Lösungen nicht mehr verlässlich arbeiten.

Details

AI system

Workflow

The police used the tools from 2020 to pre-select and classify seized image material for review. The LKA developed the tools and deployed them across Lower Saxony, and users received special training. The police trained the tools on a dataset from a criminal case that the prosecutors released for training.

Follow-up

Conditions for replication

The tools must be retrained regularly to cope with new file formats, faulty face recognition and obfuscation. A legal basis must exist for training the tools on data from criminal cases.

Entry details

Sector

Policing / child sexual abuse investigations.

Purpose

The Landeskriminalamt Niedersachsen built Kipo Analyzer and Tracebook Kipo to pre-select and classify seized image material as likely child sexual abuse material for investigators to review.

31 more questions not yet checked

model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Kipo Analyzer eases Lower Saxony police review of abuse images (AIBD-1143). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1143.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)

BibTeX Data Field definitions