6533b82ffe1ef96bd1295f60

RESEARCH PRODUCT

Artificial intelligence in the cyber security environment

Petri VähäkainuMartti Lehto

subject

cognitive abilitieskoneoppiminenanomalysupervised machine learningtekoälykyberturvallisuusunsupervised machine learning

description

Artificial Intelligence (AI) is intelligence exhibited by machines. Any system that perceives its environment and takes actions that maximize its chance of success at some goal may be defined as AI. The family of AI research is rich and varied. For example, cognitive computing is a comprehensive set of capabilities based on technologies such as deep learning, machine learning, natural language processing, reasoning and decision technologies, speech and vision technologies, human interface technologies, semantic technology, dialog and narrative generation, among other technologies. Artificial intelligence and robotics have steadily growing roles in our lives and have the potential to transform vital functions of the society. Organizations benefit from the ability of cognitive systems to improve their expertise quickly and from sharing it to all those who need it. The know-how of top experts is quickly made available to all, when their subject matter expertise is taught to a cognitive system. Through repeated use, the system will provide increasingly accurate responses, eventually eclipsing the accuracy of human experts. With artificial intelligence, comprehension can be outsourced. As the intelligence of machines improve, they will use deep learning to understand the collective information of humankind. With the use of digital sensor data, equipment based on artificial intelligence can used to develop smart advisors, teachers or assistants. As artificial intelligence technology is helping society to advance, there are risks associated with its use, found in the operating systems, hardware, algorithms, system management, ethics and liability, and privacy. The study focuses on artificial intelligence threats and risks and how AI may help to solve cyber security problems. This study uses taxonomy classification principle to classify 12 the most crucial areas of cyber security. Research method of this study was to gather 11 AI solutions that were divided into seven different categories of the crucial areas of cyber security represented in introduction chapter. AI solutions gathered uses artificial intelligence in detecting and predicting information security threats and anomalies and blocking them. The purpose of this study is to classify AI-based cyber security solutions gathered and provide information what they can offer in solving problems in the field of cyber security. peerReviewed

http://urn.fi/URN:NBN:fi:jyu-202001081077