Artificial Intelligence in Security: Using Machine Learning for Situational Awareness
Everyone is talking about artificial intelligence (AI). The way it can revolutionise business and contribute to economic growth is being described as ‘phenomenal’, with new and substantial breakthroughs being made daily. But what exactly is AI? And why should the security and intelligence industries be exploring it’s potential?
What is artificial intelligence?
Artificial intelligence, often referred to as machine learning, is a machine having the ability to not only carry out tasks that are usually reserved for humans but be able to acquire new knowledge and apply it to such tasks. This can be in the form of decision-making, visual perception and even the translation of languages.
Artificial Intelligence in Security
AI in security can provide a wealth of advantages but in order for it to work, it needs data. Which leads us to Open Source Intelligence (OSINT) and the increasing proportion of OSINT that social media and online sources is responsible for.
As of 2018, there were 500 million tweets sent every day. That’s half a billion individual pieces of data made publicly available each and every day.
Then consider how many posts are published to Facebook and Instagram every day, the number of videos uploaded to Youtube and how many messages are exchanged across the likes of Telegram and Whatsapp. Not mentioning how many news articles, reports and blogs are written on a daily basis.
That’s a lot of data.
There are already organisations within the security industry using algorithms and simple machine learning to mine the likes of Twitter and identify important information at a significant speed.
The real-time information that social media platforms, such as Twitter, broadcast can be incredibly valuable in intelligence gathering. Teamed with the inclusion of imagery and geotagging, it is a goldmine of data.
Datamining and AI enhanced technology has increased the speed in which intelligence providers identify critical information, allowing their clients to respond rapidly. Ultimately saving time, money and in some instances, lives.
But intelligence is not exclusive to Twitter. Or social media, for that matter. And mining these channels doesn’t provide context, the very element required to transform information into intelligence.
Whether you’re relying on the datamining of one channel or multiple, the staggering amount of information online can often mean that you’re receiving a lot of noise, irrelevant incidents, non-incidents, duplicate incidents and of course, fake news.
Despite the incredible potential that artificial intelligence in security possesses, can we trust it to replace the human element of analysis?
At Intelligence Fusion, we don’t believe you can substitute the value of real analysts and their talents. We do believe that machine learning is going to exponentially change the way we operate as an intelligence provider, but by enriching our service rather than becoming the sole source of it.
The development of Intelligence Fusion’s AI-enabled platform allows us to mine data from a multitude of online sources, including social media and news channels, to bring us a higher volume of data at a faster speed.
But it’s never directly uploaded onto our platform and disseminated to you as a client. What if it’s false or inaccurate information?
Intelligence that has been identified by our AI technology lands in our moderation area as a pre-populated incident. Our highly-trained and experienced analytical team verify each report, ensure the accuracy of the information and then accurately geo-locate it on our map. The incident is then approved and uploaded to our platform for our subscribers or fed to a client’s own platform via an API.
This curation process ensures that you still receive real-time threat alerts but aren’t acting on false alarms, wasting time and money, or missing out on crucial information because you’ve had to comb though numerous irrelevant incidents to get to it.
We’re also combating fake news using military standard grading techniques to monitor the credibility and reliability of every source. So you can guarantee that every piece of data pushed to you as a client has been thoroughly vetted.
Artificial intelligence and machine learning technology allows us to significantly improve our offering to clients but we can’t entirely rely on it just yet. There’s many types of incidents you won’t find on social platforms such as Twitter or Facebook, and there’s many areas of the world where the use of social media is limited or non-existent.
Volume of Twitter users across the world. Source: OmniSci Tweet Map
A list of sources that our AI mines for data.
If you need a global feed of accurate, timely and actionable intelligence, a solution entirely focused on AI and algorithms could drain manpower and money.
For intelligence you don’t have to fact-check, contact a member of our team today for your free demonstration of our threat intelligence platform. Click the button below to submit your details and a member of the team will get back to you, or email us via firstname.lastname@example.org.