Nota, a lightweight AI technology company, is expanding into the industrial and public safety sectors by unveiling video-based security monitoring technology powered by a Vision Language Model (VLM). Through 'NVA (Nota Vision Agent),' which combines existing video analysis technology with language understanding capabilities, the company presented a security monitoring method that more precisely recognizes on-site situations and detects risk factors in advance.

NOTA plans to participate in the World Security Expo (SECON 2026), held at KINTEX in Goyang, Gyeonggi Province from March 18 to 20, to introduce its technology. SECON is Korea's leading security exhibition showcasing the latest technologies in the fields of physical and cybersecurity, and is known as an annual event where officials from industry and public institutions participate to share security technology trends.

The newly unveiled NVA combines existing computer vision technology with VLM to go beyond simply identifying objects in images and is characterized by its ability to understand the context of the situation. It is designed to enhance the efficiency of the security monitoring process and the speed of response by allowing users to set detection conditions using natural language, after which the AI analyzes the situation in the video to determine the potential risk.

In addition, personal information protection and security have been enhanced by applying an on-device AI architecture that processes video data directly on the device without transmitting it to an external server. The company explained that this technology has been verified through the AI video analysis performance certification of the Korea Internet & Security Agency (KISA) and recorded an average accuracy of over 90% in the evaluation criteria.

Nota has accumulated diverse application cases by conducting security and safety-related video analysis projects in collaboration with public institutions. In fact, the technology has been utilized in urban safety management systems in partnership with local governments and for analyzing crowd density at large-scale event sites; it is also reported to have been applied by some municipalities to monitor illegal waste dumping.

At the exhibition, demos based on four scenarios will be unveiled, including early fire detection in server rooms, automatic differentiation between military and civilian vehicles and analysis of abnormal pedestrian behavior, accident risk management through crowd density analysis, and real-time detection of illegal dumping. The company believes that these capabilities can be utilized in various environments, such as industrial facilities, public infrastructure, and urban safety management.

Chae Myeong-su, CEO of Nota, explained that the company's security monitoring technology is being verified for reliability as it is being implemented in real-world settings, primarily by public institutions such as the police and local governments. He further stated that the company plans to expand the scope of application to various industrial sectors, including smart buildings, logistics, and childcare environments, based on natural language-based context understanding technology.