Abstract
Cyber security has been an emerging concern of individuals and organizations all over the globe. Although the increasing dependence of the world on the internet proves to be an advancement of technology, it also happens to be a threat to important and private information, monetary fraud leading to huge losses for organizations, and many other issues. In this day and age, cyber security has become a great necessity, and great efforts have been made to enhance it. Artificial intelligence has emerged to be of immense importance over the past decade and is expected to bear great fruit in the coming time. It is no surprise that organizations are depending on AI-driven technologies to protect their data. AI, along with concepts of machine learning and deep learning, is being used to develop new and improved means to help in the cyber security domain. The scope of this chapter covers a few artificial intelligence concepts which have been used in the past to ensure security, a few ideas which have been discussed for future implementation, threats of using AI, etc.
TopIntroduction
History of Artificial Intelligence
We owe the seeding of the idea of Artificial Intelligence to Alan Turing who first gave a mathematical approach and idea in his paper Computing Machinery and Intelligence in 1950. However, computers in 1949 could not store commands but only execute, hence there was a need to fundamentally change computers. Also, computing was extremely expensive. These were the two major factors holding back Turing from following his pursuit.
It was then a program designed by Allen Newell, Cliff Shaw, and Herbert Simon called Logic Theorist which mimicked the problem-solving skills of a human and brought the idea of Artificial Intelligence closer to reality. This program is also considered to be one of the first programs of AI.
Even though with time the understanding of Artificial Intelligence grew the main issue was still not solved, the computers were not advanced enough to carry out intelligent programs. However, with time, the computational power of machines improved along with their storage capacity to a point where Moore’s law had caught up. Moore’s law states that the number of transistors on a microchip doubles every two years and the cost of a computer reduces by half. By the 2000s AI had achieved many of its goals and this was the starting of an era where Artificial Intelligence became a field everyone had their eyes on (Haenlein and Kaplan, 2019; Kaul, Enslin, and Gross, 2020).
History of Cyber Security
During a research project ARPANET, a program called Creeper was made which moved across ARPANET’s network and left a message “I’m the creeper, catch me if you can” wherever it went. Ray Tomlinson-the inventor of email, made a program called Reaper which tracked and deleted Creeper. Reaper was the first instance of an Antivirus. By 1990s a lot of work on the antivirus part was being done and by 1992, the first antivirus program appeared. Soon as the world started coming online, virus attacks as well antivirus started getting more and more well known. Soon antivirus like McAfee came to the market, still being widely used. By the 2000s more antivirus came into the picture and now we can see an antivirus on almost every device (Warner, 2012).
Cyber Security And Ai
Principles of Cyber Security
There are certain principles followed to ensure cyber security. They are as follows:
- 1.
Confidentiality: The information should be secure and shared with authorized personnel only and no unauthorized parties should have access to the information
- 2.
Integrity: Cybersecurity to ensure that the information should be accurate, consistent and free from any modification from unauthorized organization to maintain the integrity of the information
- 3.
Availability: the cybersecurity efforts should not hinder the access of information by the authorized parties and also provide redundancy access in case of an outrage.
These are the basic principles of cybersecurity and we shall now look at the need of Artificial Intelligence in the field of cybersecurity (Štitilis, 2016).
Key Terms in this Chapter
Malware: Is a malicious software which is meant for damaging a user’s device and is often installed into devices by unsolicited emails.
Neural Networks: It is inspired by the neural networks present in the humans and later led to Artificial Neural Networks (ANNs) which are used in Deep Learning. Neural Networks rely on data to learn and improve their accuracy but once the algorithm is finely tuned, it can be used to classify data at a very high speed.
Knowledge-Based Approach: Uses several IF-ELSE conditional statements.
Machine Learning: Is a branch of Artificial Intelligence which focus on using data and applying algorithms to make predictions and classify data. Machine Learning broadly has three types of classification algorithms, supervised learning, unsupervised learning and reinforcement learning.
Supervised Learning: Has data which already has a correct answer whereas, in unsupervised learning, the algorithms cluster the data without any prior knowledge. Reinforcement learning uses a penalty system where the algorithm rewards itself for correct classification and gives a penalty for incorrect one.
Firewall: Is a network security device which is used to monitor the incoming and outgoing packets in a network. It follows a set of security rules, based on which it blocks the packets it finds suspicious.
Cyber Threat: Is on an increase especially threat to medical and financial data is severe and hence there is a need to tighten the security surrounding these fields.
Pattern-Based Approach: Data is fed into the algorithms and a pattern is noticed to carry out the algorithms.