Armenia Approves State-Backed AI Compute Program with AWS
The Government of Armenia has approved a pilot program to expand access to high-performance computing for the AI sector, delivered in partnership with Amazon Web Services (AWS). The move prioritizes practical support for teams building, training, and deploying AI systems.
The initiative runs within the Artificial Intelligence Virtual Institute, a strategic program under the Ministry of High-Tech Industry. The goal is simple: remove infrastructure barriers and accelerate credible AI work across the country's public and private sectors.
What the pilot covers
- Compute access on AWS for AI workloads including training and deployment.
- State subsidy covering 100% of actual compute usage for a limited number of selected beneficiaries.
- Technical support from AWS solution architects and access to international best practices.
Who can apply
- AI startups building products and platforms.
- Research groups and educational institutions.
- Individual specialists advancing applied AI projects.
How it works
The Artificial Intelligence Virtual Institute acts as the operational hub. It will accept applications, verify eligibility, allocate computing resources, and enable virtual collaboration across participating teams.
Selection is limited. Approved applicants receive fully subsidized compute resources on AWS, aligned with project needs and usage. The Institute also facilitates peer exchange to shorten learning curves and reduce duplicated effort.
Why this matters for government leaders
- Accelerates credible AI development without upfront infrastructure spending.
- Creates a common platform for research, industry, and public sector collaboration.
- Supports long-term ecosystem growth aligned with national AI and innovation policy.
Implementation checklist for agencies and public research bodies
- Define clear AI use cases with measurable outcomes (service delivery, research outputs, or operational efficiency).
- Prepare data governance basics: data sources, access controls, privacy posture, and audit trail.
- Identify the minimal technical stack you need (frameworks, model types, storage patterns) to avoid waste.
- Nominate a project owner and a small core team with delivery accountability.
- Line up ethical review where applicable and confirm alignment with existing policy and legal constraints.
- Plan for knowledge transfer so skills stay in-house after the pilot.
Governance and risk notes
- Clarify data classification before moving workloads to the cloud.
- Document model objectives, datasets used, and evaluation criteria.
- Track usage and outcomes from day one for accountability and later scale-up decisions.
Partners
The program is supported by AWS, with added collaboration from Mistral AI and Plug and Play. AWS brings technical expertise and architectural guidance; international partners strengthen the program's collaborative and innovation-focused components.
Key dates and how to apply
Registration opens on December 29 through the Artificial Intelligence Virtual Institute's platform at ai.gov.am. Prepare your project brief, team details, and expected resource needs before submission.
Helpful resources
- AWS High Performance Computing - overview of compute options commonly used for AI training and simulation.
- AI courses by job - useful for upskilling teams supporting pilot projects and oversight functions.
This pilot is a practical step: clear access to compute, targeted support, and a structured path to outcomes. If your team has a project that needs scale, prepare your application and move quickly.
Your membership also unlocks:
