ARM-001

Artificial Intelligence in Social Security / AI for Predictive Social Protection

Download PDF
Armenia Europe & Central Asia Upper middle income Pilot / Controlled Trial Phase Confirmed

Ministry of Labor and Social Affairs of Armenia; Nork Social Services Technology and Awareness Center

At a Glance

What it does Prediction (including forecasting) — Vulnerability, needs and risk assessment, including predictive analytics
Who runs it Ministry of Labor and Social Affairs of Armenia; Nork Social Services Technology and Awareness Center
Programme Artificial Intelligence in Social Security / AI for Predictive Social Protection
Confidence Confirmed
Deployment Status Pilot / Controlled Trial Phase
Key Risks Not assessed
Key Outcomes Enables proactive outreach—"we are no longer waiting for people to come to us, we are finding them and taking social services to them" Spdci; identifies people who do not qualify for existing programmes but have unmet needs Caucasus Watch; more accurate targeting and ability to tailor services to vulnerable citizens.
Source Quality 3 sources — News article / media, Report (multilateral / development partner)

Armenia is the an early adopter of artificial intelligence technologies in the field of social protection, according to a November 2022 presentation in Yerevan reported by Caucasus Watch. The project is implemented by the Ministry of Labor and Social Affairs of Armenia in cooperation with the Nork Social Services Technology and Awareness Center (also referred to as the Nork Information Technology Center), with funding from the Asian Development Bank (ADB). The system uses machine learning models to create mechanisms for organising social work, identifying new layers of vulnerable population groups, and providing them with more targeted support.

The initiative was first announced in January 2022, when Public Radio of Armenia reported that artificial intelligence would be used for the first time in Armenia in a system for more comprehensive and targeted assessment of beneficiary needs, rapid response, and support. The pilot test was carried out by the Nork Information Technology Center with support from ADB. The programme was designed to enable more targeted needs assessment through AI diagnostic tools and to analyse the characteristics of target groups through new technologies. By combining the capabilities of artificial intelligence, the system was intended to identify and analyse the effectiveness of the existing needs assessment system. The results of the pilot programme were to be summarised in spring 2022 and presented on international platforms.

According to Anahit Parzyan, Director General of the Nork Center, the programme was modelled over the course of one year based on the results of three years of research, after which it was presented to the Ministry of Labor and Social Affairs. Parzyan stated that the system is autonomous and works independently, and that for each new survey cycle the system can generate new analysis. She identified the main goal as analysing the needs assessment system, noting that when decisions are made and programmes for vulnerable groups are implemented, people who do not qualify for any existing programmes often remain unidentified.

The machine learning system is involved in five specific tasks, as reported by Caucasus Watch: (1) determining whether the adoption of decisions was in compliance with all legislative acts; (2) identifying beneficiaries who will remain in social security programmes for more than three years; (3) identifying beneficiaries who will exit the support programme as a result of increased income; (4) using a software algorithm to group social security beneficiaries to detect hidden trends; and (5) evaluating the effectiveness of state programmes in the field of employment. Minister of Labor and Social Affairs Narek Mkrtchyan stated that a study was conducted in cooperation with ADB to understand how AI can be helpful in social protection to make more accurate, targeted, and effective decisions.

By November 2024, the initiative had progressed to the point where Dr Anahit Parzyan, identified as Executive Director of the Nork Social Services Technology and Awareness Center, presented at the AI4SocialProtection workshop in Bangkok, co-funded by the German Federal Ministry for Economic Cooperation and Development (BMZ), the European Commission, and ADB. According to the Digital Convergence Initiative blog reporting on the workshop, Armenia is using AI to predict social protection needs in the country, leading to more accurate targeting and the ability to tailor services more closely to the needs of vulnerable citizens. Dr Parzyan stated at the workshop: 'We are no longer waiting for people to come to us, we are finding them and taking social services to them.'

Classifications follow the DCI AI Hub Taxonomy. Hover over field labels for definitions.

Social Protection Functions

Implementation/delivery chain
Assessment of needs/conditions + enrolment primaryOutreach/communications/sensitisation
SP Pillar (Primary) The social protection branch: social assistance, social insurance, or labour market programmes. Social assistance
Programme Name Artificial Intelligence in Social Security / AI for Predictive Social Protection
Programme Type The type of social protection programme, classified under social assistance, social insurance, or labour market programmes. View in glossary Other
System Level Where in the social protection system the AI is applied: policy level, programme design, or implementation/delivery chain. View in glossary Implementation/delivery chain
Programme Description An AI/ML initiative implemented by the Ministry of Labor and Social Affairs of Armenia in cooperation with the Nork Social Services Technology and Awareness Center, funded by the Asian Development Bank. The system uses machine learning to analyse the social protection needs assessment system, identify vulnerable population groups who fall outside existing programme eligibility criteria, check legislative compliance, predict long-term beneficiary status, and evaluate the effectiveness of employment programmes.
Implementation Type How the AI output is produced: Classical ML, Deep learning, Foundation model, or Hybrid. Affects validation, compute requirements, and governance profile. View in glossary Classical ML
Lifecycle Stage Current stage in the AI lifecycle, from problem identification through to monitoring, maintenance and decommissioning. View in glossary Integration and Deployment
Model Provenance Origin of the AI model: developed in-house, adapted from open-source, commercial/proprietary, or accessed via third-party API. View in glossary Not documented
Compute Environment Where the AI system runs: on-premise, government cloud, commercial cloud, or edge/device. View in glossary Not documented
Sovereignty Quadrant Classification of data and compute sovereignty: I (Sovereign), II (Federated/Hybrid), III (Cloud with safeguards), or IV (Shared Innovation Zone). View in glossary Not assessed
Data Residency Where the data used by the AI system is stored: domestic, regional, or international. View in glossary Not documented
Cross-Border Transfer Whether data crosses national borders, and if so, whether documented safeguards are in place. View in glossary Not documented
Decision Criticality The rights impact of the decision the AI supports. High criticality requires HITL oversight; moderate requires HOTL; low may operate HOOTL. View in glossary High
Human Oversight Type Level of human involvement: Human-in-the-Loop (active review), Human-on-the-Loop (monitoring), or Human-out-of-the-Loop (periodic audit). View in glossary HOTL
Development Process Whether the AI system was developed fully in-house, through a mix of in-house and third-party, or fully by an external provider. View in glossary Not documented
Highest Risk Category The most significant structural risk source identified: data, model, operational, governance, or market/sovereignty risks. View in glossary Not assessed
Risk Assessment Status Whether a formal risk assessment, informal assessment, or independent audit has been conducted for this system. Not assessed

Risk Dimensions

Governance and institutional oversight risks
Operational and system integration risks

Impact Dimensions

Autonomy, human dignity and due process
CategorySensitivityCross-System LinkageAvailabilityKey Constraints
Beneficiary registries and MISPersonalLinks data across multiple systemsCurrently available and usedData drawn from 15+ social protection information systems; requires integration across multiple beneficiary databases managed by Nork Center
Survey and census dataPersonalSingle source (no linkage)Currently available and usedNeeds assessment survey data collected periodically; system generates new analysis for each new survey cycle

Caucasus Watch (2022) 'Armenia Becomes First in World to Apply AI in Social Protection', Caucasus Watch, November 2022. Available at: https://caucasuswatch.de/en/news/armenia-becomes-first-in-world-to-apply-ai-in-social-protection.html (Accessed: 23 March 2026).

View source News article / media

Digital Convergence Initiative (2024) 'AI in Social Protection – Now and Tomorrow', SPDCI Blog, November 2024. Available at: https://spdci.org/resources/blog-ai-in-social-protection-now-and-tomorrow/ (Accessed: 23 March 2026).

View source Report (multilateral / development partner)

Public Radio of Armenia (2022) 'Armenia to Use AI for Needs Assessment and Support Programs', Public Radio of Armenia, 31 January 2022. Available at: https://en.armradio.am/2022/01/31/armenia-to-use-ai-for-comprehensive-need-assessment-and-support-programs/ (Accessed: 23 March 2026).

View source News article / media
Deployment Status How far the system has progressed into real-world operational use, from concept/exploration through to scaled and institutionalised. View in glossary Pilot / Controlled Trial Phase
Year Initiated The year the AI system was first initiated or development began. 2022
Scale / Coverage The scale and geographic or population coverage of the deployment. Unknown
Funding Source The source(s) of funding for the AI system development and deployment. Unknown
Technical Partners External technology vendors, academic partners, or development partners involved. In-house development by Nork Social Services Technology and Awareness Center
Outcomes / Results Enables proactive outreach—"we are no longer waiting for people to come to us, we are finding them and taking social services to them" Spdci; identifies people who do not qualify for existing programmes but have unmet needs Caucasus Watch; more accurate targeting and ability to tailor services to vulnerable citizens

How to Cite

DCI AI Hub (2026). 'Artificial Intelligence in Social Security / AI for Predictive Social Protection', AI Hub AI Tracker, case ARM-001. Digital Convergence Initiative. Available at: https://socialprotectionai.org/use-case/ARM-001 [Accessed: 1 April 2026].

Change History

Updated 31 Mar 2026, 06:35
by system (system)
Created 30 Mar 2026, 08:38
by v2-import (import)