AI in video-based security
The report highlights current developments, key opportunities, future trends and risks for AI in video-based security
Axis Communications, a technology leader in network video, has published its first AI report, ‘The State of AI in Video Surveillance 2025.’ It highlights current perspectives for the use of artificial intelligence (AI) in video-based security and beyond. The report offers valuable insights into AI technologies, their integration and the associated opportunities, challenges and risks – particularly with regard to security, operational efficiency and business intelligence.
The use of AI has increased significantly over the past two years, mainly due to rising customer demands, greater knowledge and experience with potential applications, and newly developed application scenarios. Axis therefore interviewed AI experts in its global partner network about the importance of AI in their respective industries and compiled and analysed the insights gained on AI technologies in its new report.
The results at a glance:
- Executives: 66 percent of executives surveyed consider AI and generative AI to be the most important priority for their company in the near future, followed by cybersecurity, cyber risks and data protection (61 percent) and analytics and actionable insights (56 percent).
- Partners: 62 percent of the partners surveyed rate AI and generative AI as the most important future trend, on a par with cybersecurity, cyber risks and data protection (also 62 percent), followed by analytics and actionable insights (50 percent).
- End customers: For end customers, cybersecurity, cyber risks and data protection (61 percent) are the top priority, followed by analytics and actionable insights (36 percent) and AI and generative AI in third place (34 percent).
The shift to cloud and edge AI continues
As the report shows, the shift from locally operated server systems to hybrid architectures continues to progress rapidly. This development is driven by a need for greater scalability, faster data processing and optimised bandwidth utilisation. A hybrid model that combines the ability of edge AI – i.e. AI-based real-time analysis ‘on the edge’ directly on the cameras themselves – for immediate data processing with the scalability and long-term data storage in the cloud is becoming the preferred approach for many companies. This flexibility allows them to leverage the advantages of both technologies.
The integration of different data sources
Respondents agreed that the integration of additional sensor data, such as audio and context-related environmental data and video data, is a useful addition and leads to more comprehensive and actionable insights. This enables more accurate analysis and assessment of events and situations, which improves both security and business intelligence functions, i.e. functions for optimising the entire business process.
Combining multiple data streams enables more accurate detection and prediction of potential threats and anomalies. By linking video and audio analysis data, security managers can respond more quickly and effectively in emergency situations. This more comprehensive, context-sensitive approach mimics the way humans naturally absorb and process information from various data sources in their environment, representing an important step forward in improving security and business processes.
New application scenarios for operational efficiency and business intelligence
The integration of AI-based analytics capabilities turns network cameras into powerful, data-generating sensors. Cloud computing provides AI-enabled edge devices such as cameras with the necessary computing power and can redefine the video security industry. In the interviews, respondents highlighted this paradigm shift and also emphasised that AI-enabled video solutions are now also being used in industries such as retail, logistics and manufacturing.
Common application scenarios in these areas include optimising space planning, people counting, asset tracking and resource management. One topic that comes up repeatedly in this context is the potential for AI to deliver significant cost savings and improvements in operational efficiency. Embedding video analytics in a broader context of business intelligence systems can help companies improve resource allocation and process automation and enable more informed decision-making.
Face recognition is gaining importance
The report also shows that facial recognition is increasingly being used in various countries. This is being driven by new regulatory requirements that provide greater clarity on ethically acceptable use cases for the technology and a framework for its responsible use.
Respondents expect facial recognition to become even more important worldwide. But before that happens, data protection regulations must be adapted and clarity created about how this technology works. Ethical issues therefore remain a focus, especially in regions with strict data protection laws.
‘Artificial intelligence is one of the most powerful and transformative technologies in video-based security today. Our report highlights that AI offers tremendous opportunities to improve security, increase operational efficiency and optimise business processes in the long term. At the same time, companies in our region must continue to place great emphasis on ethical implementation and meaningful integration in order to create sustainable value,’ says Tobias Metsch, Regional Director Middle Europe at Axis Communications.
Method
The report combines insights from in-depth interviews with AI experts from Axis Communications’ global partner network, conducted throughout 2024, with data from two global surveys conducted by Axis Communications among approximately 4,900 distributors and sales partners in 68 countries and approximately 900 end customers in 64 countries in the second quarter of 2024.
