Artificial Intelligence, or AI, is something of a buzzword of late, but it is a very powerful tool in our digital age. One of its most important roles is security - both in our digital and real-world lives.
Here we will briefly explore where it is being used, how it is being used, and provide some current interesting examples.
How is AI being used in security?
AI and machine learning are being employed increasingly around the world to help preserve and improve security in many ways. From helping keep ahead of the fast-paced development of cybersecurity threats to helping law enforcement and security services prevent criminal acts, AI is becoming an essential tool to keep us all safe from malicious, or even dangerous, neer-do-wells.
For cybersecurity, the role of AI is more intuitive for the non-initiated. Cyber attacks, in particular, are growing in complexity and volume around the world.
Here AI can be employed to help, often, under-resourced security operations analysts stay ahead of the curve. AI can, for example, curate all current knowledge of threat intelligence to help provide threat insights almost instantaneously.
This helps reduce response times to cyberattacks considerably. AI can also be trained to learn by consuming billions of data artifacts from structured and unstructured sources.
This could include blogs and news stories and allows the AI, using machine learning, to improve its knowledge of cybersecurity over time. More sophisticated ones, like IBM's Watson, can even employ a form of cyber-reasoning to find relationships between suspicious files or IPs in seconds to minutes.
This greatly improves cybersecurity analysts response times to potential threats.
What companies are working on artificial intelligence security?
There are various companies working on AI security. Many larger organizations, like Microsoft and IBM, also have AI-security departments.
As we previously mentioned, AI can also help with physical security. Companies like Liberty Defense, a concealed weapon detection company, have been working on an AI-powered solution to help reduce weapon-related crimes.
Called HEXWARE, the weapons detection system uses active 3D imaging and AI to assess and detect threats as groups of people pass its sensors. It was developed at MIT's Lincoln Laboratory can have been designed to be installed at the perimeter of any building.
It can be used both indoors and outdoors and detect both metallic and non-metallic threats. This is one of the few technologies of its kind that doesn't also use facial recognition - which protects the privacy of individuals.
The idea is to allow a site's security to identify a possible threat before it even enters the building.
Another example is OPENALPR's AI security software. It uses IP cameras to scan license plates and provide vehicle data including tag number, make, model, and color in real time. It recently announced it saw a 2,960% increase in coverage over the past two years, and now has 9,000 camera installs in over 70 countries.
This has wide-ranging applications from helping law enforcement detect a pattern in criminal behavior to more benign issues like parking management.
What is cognitive security?
Cognitive security is defined as:
Like many other cognitive computing techniques, these self-learning systems use data mining, pattern recognition, and natural language processing to simulate the human brain.
According to IBM, Cognitive Security can be characterized as:
"Cognitive computing [or security], an advanced type of artificial intelligences, leverages various forms of AI, including machine-learning algorithms and deep-learning networks, that get stronger and smarter over time."
This kind of cybersecurity is very effective at helping to stop cyber attacks designed to manipulate human perception.