'Deepfake' and artificial intelligence (AI)-based disinformation should now be considered part of the hybrid threat arsenal, alongside traditional cyberattacks, Chairman of the Temporary Commission against Foreign Interference and Hybrid Threats of the Azerbaijani Parliament, Ramid Namazov, said, Trend's correspondent reports from the event.
The MP made the remark during a public debate in the parliament on the topic "Digital development, artificial intelligence and protecting children from harmful effects on social network platforms."
He noted that the National Supercomputer Center has already been established in the country, which allows complex AI models to be safely trained and managed within the country, without depending on external cloud services.
According to him, this is a strategic step in terms of both national security and digital sovereignty, and sensitive state information can now be processed without leaving the country's borders.
Namazov pointed out that the development of AI is not only of strategic importance in terms of digital independence, information security, and national security, but also creates broad opportunities for increasing efficiency in public administration, optimizing decision-making processes, and diversifying the economy.
The official emphasized that against the background of threats created by AI, the focus is shifting to not only cyber defense, but also the concept of digital resilience, which involves rapid recovery after cyber incidents. He noted that the country's growing digital dependence and complex geopolitical position require a fundamental strengthening of the cybersecurity strategy, the training of qualified personnel, and the implementation of risk management mechanisms at the national level.
Namazov highlighted that one of the newest and most sophisticated threats created by AI is deepfake content and synthetic media, called "deepfake". According to him, artificial intelligence solutions are now able to create such realistic and fully functional fake audio and video materials that the speed of their spread sometimes exceeds the ability of state agencies or fact-checking mechanisms to react.
He said that this is not just a technological phenomenon, but has become one of the most agile and difficult-to-track tools of modern hybrid threats. Namazov believes that the 'deepfake' and AI-based disinformation should be assessed alongside traditional cyberattacks, taking into account that these tools are used to undermine public trust, damage the reputation of state institutions, as well as deceive and manipulate individuals and children.
According to him, the risks associated with disinformation and artificial intelligence can also be assessed as a separate category of threats. Namazov noted that the systems for automatic detection of prohibited content through AI, envisaged within the framework of the 2026–2028 Action Plan, are aimed at responding to these new generation threats.
The commission head said that the next important step is the creation of large language models in the Azerbaijani language by the end of 2028. According to him, since most of the existing large language models in the world are trained on English and several dominant languages, high-quality AI services in the Azerbaijani language are limited. He emphasized that the creation of strong models in the Azerbaijani language will further strengthen cultural and linguistic sovereignty.
Namazov pointed out that the AI Strategy covers four main areas. These are improving the management of AI, developing data management and computing infrastructure, strengthening human resources, as well as public education.
According to him, the strategy identifies informing society not only about the possibilities of AI, but also about its risks and protecting ethical principles as one of the state's priority tasks. The AI Academy, established within this framework, aims to train 500 engineers through programs in line with international standards and form an AI community of 3,000 people.
Namazov added that Azerbaijan's AI, benefits from international partnerships in other areas.
