Skip to main content
AI Hub
Home Browse Cases Countries Sources Explore Taxonomy About Submit
Sign In
DCI AI Hub — AI Tracker socialprotectionai.org/use-case/COD-001
COD-001 Exported 1 April 2026

WFP SCOPE – Biometric Verification at Food/Cash Distributions (DRC)

Country Congo, Dem. Rep.
Deployment Status Operational Deployment (Limited Rollout)
Confidence Confirmed
Implementing Agency World Food Programme (WFP) – DRC Country Office; cooperating partners (e.g., CARITAS agents trained for biometric registration).

Overview

WFP SCOPE is the World Food Programme's beneficiary information and transfer management platform, described by WFP as an in-house developed technology that functions as a database to securely store beneficiary information and manage the transfer of benefits. In the Democratic Republic of Congo (DRC), SCOPE is deployed for biometric verification during food and cash distributions to ensure that entitlements reach the correct beneficiaries and to prevent duplication at the beneficiary level. The DRC Country Office was one of nine country offices sampled in WFP's 2021 global internal audit of SCOPE (AR/21/08), and was the subject of a dedicated internal audit of WFP operations (AR/20/12) covering the period January to December 2019.

The biometric deduplication capability within SCOPE allows operators to identify individuals who may be registered under multiple different identities, investigate potential duplicates, and deactivate confirmed duplicated identities. According to the SCOPE User Manual, 'Deduplication ensures effective programming and is critical to ensuring that WFP and partner funds reach targeted beneficiaries as intended without duplication at the beneficiary level.' Biometric information including fingerprints and face scans are used to perform biometric deduplication. The system also supports deduplication through data analytics without the use of biometrics, as biometrics may not always be appropriate or necessary in some programmes.

The biometric deduplication process consists of several stages. First, automatic deduplication occurs when new identities are registered in SCOPE with biometric data. Their biometric information is automatically checked by the MegaMatcher Automated Biometric Identification System (ABIS). The ABIS analyses the registered biometric data for new identities and compares it against all existing registered identities in SCOPE, assigning match scores to determine whether an identity is duplicate, unique, or requires manual adjudication. Identities with a match score below 96 are automatically flagged as unique. Identities with match scores between 96 and 144 are flagged for manual adjudication, as the ABIS is unable to determine with certainty whether they represent duplicates. Identities with match scores above 144 are automatically flagged as duplicates. The SCOPE User Manual notes that as MegaMatcher ABIS is further optimised, these match score ranges may change.

For identities flagged for manual adjudication, human adjudicators must manually review and determine whether the identities are genuine duplicates or distinct individuals. This human-in-the-loop process provides an important safeguard against both false positive and false negative deduplication decisions. Identities confirmed as duplicates can be downloaded as a duplicated identities list for field investigation by programme staff. Once duplicates are confirmed, the duplicated identities are deactivated so that only the individual's actual identity remains active in SCOPE.

At the global level, SCOPE had been implemented in 68 of the 85 countries where WFP has a presence by 2020, with almost 63.8 million identities registered and 20.2 million beneficiaries actively managed. A total of 3.2 million SCOPECARDs had been used globally, with PIN (32.5 percent), fingerprints (58.4 percent), or iris, QR code, or beneficiary photo (9.2 percent) as the main verification forms. SCOPE also offers a Real-Time Biometric Identification tool that allows country offices to check for duplicate records at the time of registration, which can be used both offline and online.

In the DRC specifically, the 2020 internal audit (AR/20/12) found that 2.65 million cash-based transfer beneficiaries were recorded in the DRC SCOPE platform as of February 2020, though the audit assessed that less than 50 percent of these could actually be considered active. The bulk of beneficiary data dated back to a 2016 migration exercise, where data was not properly reviewed, checked for completeness, or validated prior to its migration into SCOPE. The audit identified significant data quality issues, including 11,000 duplicate households, 1,949 duplicate electoral cards, 2,050 duplicate individuals, and 25,000 duplicated fingerprints. Eighty percent of the sampled duplicated fingerprints related either to individuals registered under different names, or to different individuals with different names and identification signs.

The DRC audit further noted that SCOPE was used only for the cash-based transfer modality, with biometric verification available only in certain locations. In locations where biometric verification was not available, the verification process was manual and performed jointly by WFP, cooperating partner, and financial service provider representatives. The Country Office did not use SCOPE for beneficiary data management of in-kind assistance, relying on Excel files instead. At the time of audit reporting, the Country Office had procured and was rolling out real-time data de-duplication software, and its data management and improvement plan had resulted in the deactivation of over one million of the 2.65 million beneficiaries in SCOPE. No privacy impact assessment had been carried out for the use of beneficiary data for either cash-based transfer or in-kind activities at the time of the 2020 audit.

Classification

AI Capabilities

Classification (primary)Perception and extraction from unstructured inputs

Use Cases

Identification, verification and record linkage (primary)

Social Protection Functions

Implementation/delivery chain: Registration (primary)
SP Pillar (Primary)Social assistance

Programme Details

Programme NameWFP SCOPE – Biometric Verification at Food/Cash Distributions (DRC)
Programme TypeIn-Kind Transfers
System LevelImplementation/delivery chain

WFP food and cash distribution programme in the Democratic Republic of Congo, using the SCOPE platform for beneficiary registration, biometric identity verification, and deduplication to ensure assistance reaches targeted beneficiaries without duplication.

Implementation Details

Implementation TypeDeep learning
Lifecycle StageMonitoring, Maintenance and Decommissioning
Model ProvenanceNot documented
Compute EnvironmentNot documented
Sovereignty QuadrantNot assessed
Data ResidencyNot documented
Cross-Border TransferNot documented

Risk & Oversight

Decision CriticalityHigh
Human OversightHITL
Development ProcessNot documented
Highest Risk CategoryNot assessed
Risk Assessment StatusNot assessed

Risk Dimensions

Data-related risks

Consent or lawful basis gapData quality failureRepresentation biasWeak provenance or lineage

Governance and institutional oversight risks

Insufficient institutional capacityWeak documentation or auditability

Model-related risks

Reliability or generalisation failure

Operational and system integration risks

Monitoring gap

Impact Dimensions

Equality, non-discrimination, fairness and inclusion

Systematic exclusion from benefits or services

Privacy and data security

Disproportionate surveillance or profilingLoss of individual control over personal dataPrivacy violation or data breach

Safeguards

Human oversight protocol

Deployment & Outcomes

Deployment StatusOperational Deployment (Limited Rollout)
Year Initiated2016
Scale / CoverageUnknown
Funding SourceUnknown
Technical PartnersWFP SCOPE beneficiary & transfer management platform (supports registration, deduplication, authentication, and distribution tracking).

Outcomes / Results

WFP SCOPE biometric deduplication and verification at distribution points reduce duplicate beneficiary records and impersonation risk during aid delivery operations.

Sources

  1. SRC-004-COD-001 United Nations Global Marketplace (2021) 'Provision of Contactless Biometrics Solutions for Iris Capture', UNGM Notice 118658. Available at: https://www.ungm.org/Public/Notice/118658 (Accessed: 30 March 2026).
    https://www.ungm.org/Public/Notice/118658
  2. SRC-001-COD-001 World Food Programme (n.d.). Biometric Deduplication (ABIS) – SCOPE User Manual. Rome: WFP. Available at: https://usermanual.scope.wfp.org/cash-accounts/content/intros_to_sections/deduplication.htm (Accessed 30 Oct 2025).
    https://usermanual.scope.wfp.org/cash-accounts/content/intros_to_sections/deduplication.htm
  3. SRC-002-COD-001 World Food Programme, Office of the Inspector General (2020). Internal Audit of WFP Operations in the Democratic Republic of Congo. Internal Audit Report AR/20/12. Rome: WFP. Available at: https://docs.wfp.org/api/documents/WFP-0000118040/download/ (Accessed 23 Mar 2026).
    https://docs.wfp.org/api/documents/WFP-0000118040/download/
  4. SRC-003-COD-001 World Food Programme, Office of the Inspector General (2021). Internal Audit of SCOPE: WFP's Digital Management of Beneficiaries. Draft Internal Audit Report AR/21/08. Rome: WFP. Available at: https://docs.wfp.org/api/documents/WFP-0000128891/download/ (Accessed 23 Mar 2026).
    https://docs.wfp.org/api/documents/WFP-0000128891/download/

How to Cite

DCI AI Hub (2026). 'WFP SCOPE – Biometric Verification at Food/Cash Distributions (DRC)', AI Hub AI Tracker, case COD-001. Digital Convergence Initiative. Available at: https://socialprotectionai.org/use-case/COD-001

Back to case page
AI Hub

Digital Convergence Initiative - AI Hub

Responsible, ethical use of AI in social protection

MarketImpact Platform developed by MarketImpact Digital Solutions
Co-funded by European Union and German Cooperation. Coordinated by GIZ, ILO, The World Bank, Expertise France, and FIAP.