AI trends in video surveillance focus on ethics, hybrid models

Axis Communications has published a study examining the evolving role of artificial intelligence in the video surveillance sector, with a focus on ethical considerations and industry trends.

The report, titled 'The state of AI in video surveillance' (2025), analyses how AI is transforming security and operational efficiency by enabling the extraction of business intelligence from video data. The study draws on insights from 11 expert interviews and the results of two global surveys conducted among more than 5,500 respondents across over 60 countries in 2024.

According to the report, demand for AI-driven surveillance solutions has increased substantially in recent years, prompted by customer interest, broader awareness of applications, and the appearance of new use cases. However, the deployment of these technologies is still maturing, with evolving best practices and implementation processes.

The research identifies several key trends, including a growing emphasis on ethical approaches, data protection, and regulatory compliance as fundamental requirements for the sector's future. The report also notes the necessity for greater collaboration through partnerships to address these challenges.

Mats Thulin, Director AI & Analytics Solutions at Axis Communications, commented: "AI remains one of the most powerful and transformative technologies within the video surveillance industry. This new research reveals that while there are significant opportunities for AI to improve safety and security, operational efficiency and business intelligence, there must be a focus on ethical implementation and meaningful integrations which drive value."

The application of AI extends beyond traditional security measures, with intelligent cameras now being used to enhance efficiency and aid decision-making in various sectors. The transition to hybrid architectures combining cloud and edge AI is accelerating, supporting the need for local analytics and rapid responsiveness during critical incidents.

Survey participants highlighted the primary trends shaping the industry, including concerns over cybersecurity, risk and privacy (61%), analytics and actionable insights (36%), and the advent of AI and generative AI (34%). These findings illustrate both the business and societal benefits of AI—such as improved operational efficiency, better decision-making, and higher output quality—as well as the associated challenges.

The report outlines nine thematic insights from expert interviews. Among these, the migration to hybrid cloud and edge AI models stands out. The hybrid approach is gaining favour, as it leverages the real-time processing capabilities of edge AI on devices, alongside the scalability and storage capacity provided by cloud infrastructure.

Another finding is the significance of integrating diverse data sources. By combining video with data from additional sensors—such as audio and environmental inputs—organisations can achieve a more comprehensive view, enhancing situational awareness and generating deeper, more actionable insights.

Facial recognition technology is reported to be gaining momentum worldwide, but the report underscores the need for strict adherence to privacy regulations and greater transparency about the use and operation of these systems. Ethical considerations regarding facial recognition are highlighted as a central focus, particularly in regions with robust privacy laws.

The report also emphasises the importance of embedding responsible AI practices from early development stages. This involves designing systems that prioritise fairness, transparency, and data privacy from inception rather than as afterthoughts. Human oversight is identified as crucial for responsible AI deployment, with transparency viewed as essential for building trust.

Thulin further stated: "As a company, Axis is committed to the advancement of responsible AI. Our approach to AI is rooted in one overarching principle: That AI technology, just like all technologies, should leverage and augment human intelligence, build on respect for human rights and should benefit people and society. This is in line with Axis vision, to innovate for a smarter, safer world."

The methodology behind the study involved semi-structured interviews with technology partners and system integrators who possess extensive AI expertise, as well as analysis of responses from channel partners and end customers across several continents. Key themes of ethical AI, the move toward hybrid cloud-edge environments, collaborative integration, and responsible application are highlighted through the gathered data and expert opinions.