Skip to main content

Ինտերնետ հասարակության Հայաստան հատված

Could AI Become a Tool of State Surveillance? Risks for Armenia and the Region

Artificial intelligence can be used both to improve public services and to monitor people and restrict online content. This was discussed at the ArmIGF2026 conference by Dmitry Kuteinikov, General Counsel at RKS Global and a researcher in AI governance and regulation.

His research, Human versus Model, examines how nine countries – Armenia, Georgia, Russia, Belarus, Kazakhstan, Kyrgyzstan, Tajikistan, Uzbekistan, and Azerbaijan – use AI for content moderation and population management.

According to the research, the adoption of technology is progressing faster than the development of legal frameworks to regulate it. The countries examined are at different stages of AI regulation. Kazakhstan and Kyrgyzstan have adopted AI-related laws, while Russia’s new law entered into force in September 2026. Azerbaijan and Uzbekistan have introduced certain regulatory measures, and Tajikistan has an AI development strategy. Armenia, Belarus, and Georgia, however, still lack dedicated AI legislation or a detailed national strategy.

One of Kuteinikov’s main concerns is that existing regulations often do not extend to security and law enforcement agencies. At the same time, facial recognition systems are already being used or introduced in all the countries covered by the research. In many cases, these technologies are deployed before clear rules governing their use are established and without adequate independent oversight.

The research also shows that restricting online content is an available tool in all the countries examined. According to the report, Armenia and Georgia block websites more selectively than the other countries. Meanwhile, seven countries have already developed or are developing their own AI models. In this context, it is important to understand who determines what content these systems can generate and what information will be accessible to users.

With regard to Armenia, the study identifies three main areas of concern:

  • AI regulation;
  • Surveillance technologies;
  • Personal identification proposals.

Armenia has signed, but has not yet ratified, the Council of Europe’s Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law. The national AI strategy is still under development, and there is no dedicated AI law yet. According to the research, since 2025, the police have been able to use public surveillance cameras for biometric identification.

Particular attention was given to the proposal to establish a unified registry of mobile device IMEI numbers. The draft proposes bringing together device identifiers, user information, SIM and eSIM card details, as well as border-crossing and customs data. The researcher noted that the draft does not specify a data retention period. He also referred to a proposal introduced in August 2026 that would require identity verification when creating social media accounts.

According to Kuteinikov, combining these datasets could link individuals to their identity documents, mobile phones, and online accounts, increasing the risks of personal data centralisation and potential misuse. He stressed that online anonymity is an essential component of freedom of expression, while mandatory identification would not solve the problem of fake accounts operated from abroad.

In light of these concerns, RKS Global sent an open letter to the Government of Armenia and the National Assembly in October, recommending against the introduction of universal mandatory identification for internet users. The organisation proposes that a user’s identity should only be disclosed in specific cases, following a court order and an assessment of proportionality. It also recommends publishing a Data Protection Impact Assessment before establishing the IMEI registry.

Based on the research findings, Kuteinikov emphasised the importance of ensuring that AI development is accompanied by clear legislation, independent oversight, and safeguards for the protection of human rights. He also stressed that public authorities and private companies should operate under the same principles and that the use of high-risk AI systems should be subject to strict oversight.

Facebook
Twitter
LinkedIn
Գլխավոր_result
Artificial Intelligence Makes Work Easier, but It Cannot Replace Humans
Գլխավոր_result
Developing a Digital Armenia Requires Modernizing Connectivity Infrastructure and Ensuring Equal Access
Գլխավոր_result_
Security of Critical Infrastructure Must Be Built in from the Design Stage
Գլխավոր_result
AI Systems Must Be Transparent, Explainable, and Meet Data Quality Requirements
Գլխավոր_result
From Artificial Intelligence to Digital Trust: Internet Development Challenges Discussed at ArmIGF2026
ArmIGF պատմություն_result
Shaping the Future of the Internet Through Cooperation: The IGF Brings Everyone Together
C297B809-C851-4941-948A-A6948B2F631E_result
Multistakeholder Governance: Who Decides the Future of the Internet and How?
A5CA7323-048A-4319-8FC7-771685B439D8_result
A Unified and Connected Internet: From Global Connectivity to Armenia’s Digital Opportunities
Էլեկտրամատակարարում_20261028_155359_0001_result
Critical Infrastructures: System Resilience and National Security
1000017745_result
Protecting Digital Data Through Encryption
Subscribe for our newsletter.