AI (Artificial Intelligence) is a really hot topic at the moment, not just in security but in the wider field of technology. Products we use every day (from smartphones and smart-speakers to cars) are boasting the use of AI technology to aid their functions more efficiently.
There is some debate over whether all these systems use AI in the truest sense (which is not something I will argue here!) but needless to say, society is becoming very used to the idea of machines making decisions to aid us in our daily lives. It will come as no surprise that the security industry has been embracing the concept wholeheartedly for some time now.
With the Internet of Things (IoT) becoming widely adopted there is a huge amount of extra data being generated, as previously siloed systems suddenly become part of the business facility or home network. The figures are astounding – research by IT giant Cisco suggests 5 quintillion bytes of data are produced every day worldwide already, so increased IoT systems will enlarge this even further with estimates suggesting there will be more than 30 billion connected devices by 2020!
Undoubtedly, as sensors and general IoT-enabled devices become cheaper, these figures will grow exponentially. We’ll be able to know a huge amount of information about any connected building, from heat/humidity to where people are and how they interact with their environment. This is a wealth of valuable data that simply can’t be digested and analysed by humans alone.
Taking this raw data and gaining valuable insights from it is an opportunity that is easily missed, but one which is well-worth grasping if possible! Obviously, it’s perfect for detecting security issues and enhancing the protection of the facility, but it’s also an excellent way to make a building more efficient and to improve the efficiency of the organisation as a whole. This is where the power of AI can make a really striking difference.
It doesn’t matter if the AI is examining the actions of a potential intruder or the day-to-day movement of staff and visitors – it examines the data and makes recommendations (or direct decisions if instructed to do so) from looking at a broader picture of data.
AI in Use
CCTV and Video Management Systems (VMS) have been enjoying the benefit of AI assistance for a few years now. Traditionally this sort of approach involved the data being processed by a central core system – so various feeds from CCTV would be fed to a powerful processor to react and call for action as required. The issue with this approach was that any interference with the network could delay or block responsive action.
To get around this, the best security systems now employ AI at the edge of the network – with processing done by the systems that are directly monitoring the situation all the time. This is not only reliable but also ensures there are no delays or interruptions.
Inevitably there has always been a level of reservation over the use of AI – a common worry being it will replace people and their jobs. In fact, what AI in security does is enhance and empower human teams.
In the past, a CCTV operator (for example) would be expected to watch a small screen to look for issues. Realistically the human brain (and attention span) is not designed to just watch a feed and maintain full alertness throughout. An AI system is far better at doing this unhindered.
Equally, whilst it is perfect at spotting potential issues, an AI system will need physical backup to investigate a situation and act if required. Using AI on the edge of the network means alarms and suspicious activity can be monitored automatically and the human team left alone to complete more suitable and less laborious tasks. This also means the human team resources aren’t wasted, so the overall effectiveness is increased and made more acute.
Embracing the Technology
Whilst the results are undoubtedly impressive, we are a long way off from AI being the complete security and facilities management panacea that some imagine. Also, would or could you ever completely rely on AI to adequately deal with the complex moral and ethical dilemmas that a human security team must cope with?
At the end of the day, what is needed is a way to enhance the human security and management teams so they can oversee vast quantities of surveillance data and act accordingly. This is the sweet spot of modern AI systems.
Think of it as an electronic assistant, one which can process information and show you what you need to see without troubling you with everything else – the perfect combination of people and machines cooperation.