Initiative file / Fiscal

Electronic-invoicing anomaly detection

Evidence record v2

14 / 31

Institution
Finance
Editorial classification
Verified adoption
Initiative type
AI system
Documented phase
Operational
Relationship to AI in the source
Explicit relationship to AI
Execution evidence
Confirmed
First evidence
Last verified

Documented scope

What it is

Finance Ministry solution for detecting anomalous patterns in electronic invoices and prioritizing audits. The project began in 2022, was tested in 2023, and the 2025 Macrofiscal Risk Report records continued use of the E-IAD detector.

Objective stated by the institution

Detect anomalous patterns in electronic invoices to prioritize tax-audit cases.

Known timeline

Only milestones with an explicit date in the sources or editorial review are shown.

  1. First evidence
  2. Pilot start
  3. Operational start
  4. Latest editorial verification

Evidence reading

Each dimension is assessed separately. Undetermined marks an evidence gap; not applicable means the dimension does not correspond to the documented object.

What is confirmed

  • The Finance Ministry records continued E-IAD use in 2025; CIAT documents its installation in Costa Rica and its unsupervised machine-learning technique.
  • The news report states that human technical review occurs before audit actions begin.

What could not be determined

  • No independent evaluation of accuracy or false positives was found.

Documented results

Results are published only when a source traceably attributes them to this initiative.

  1. The Finance Ministry reported 50 audited cases and approximately ₡8 billion recovered during 2025.

    [4]

Evidence matrix

Existence

Confirmed

Execution

Confirmed

AI technique

Confirmed

Operational AI use

Confirmed

Results

Confirmed

Governance

Confirmed

Sources and traceability

Sources state which claims they support, their origin and the date on which they were consulted.

  1. [1]
    2025 Macrofiscal Risk Report
    Publisher
    Ministerio de Hacienda
    Source type
    Official primary source
    Published
    Consulted

    Supports: existence · execution · operational use

  2. [2]
    E-IAD: Electronic Invoicing Anomaly Detector
    Publisher
    Centro Interamericano de Administraciones Tributarias
    Source type
    Multilateral organization
    Consulted

    Supports: existence · execution · AI technique · operational use

  3. [3]
    Costa Rica and CIAT deploy an e-invoicing anomaly-detection solution
    Publisher
    Microsoft
    Source type
    Other secondary source
    Published
    Consulted

    Supports: existence · execution · AI technique · operational use

  4. [4]
    Finance Ministry explains the technology behind ₡8 billion in recoveries
    Publisher
    La Nación
    Source type
    News media
    Published
    Consulted

    Supports: existence · execution · AI technique · operational use · reported result · governance

Open questions

No additional open questions were recorded in the current review.

Documented relationships

Context

This entry covers the electronic-invoicing anomaly-detection solution. The Finance Ministry confirms continued use, and CIAT documents installation and unsupervised machine learning. TRIBU-CR and ATENA are distinct systems.

Related projects