ISSN Approved Journal || eISSN: 2582-8185 || CODEN: IJSRO2 || Impact Factor 8.2 || Google Scholar and CrossRef Indexed
Fast Publication within 48 hours || Low Article Processing Charges || Peer Reviewed and Referred Journal || Free Certificate
Adversarial Attacks on AI Systems: A Growing Cyber Threat
1 Masters in Computer Science, Louisiana State University in Shreveport.
2 Computer Science and Engineering, Stamford University Bangladesh.
3 Department of Computer Science, American International University-Bangladesh.
4 Department of Computer Science and Engineering, Daffodil International University Dhaka Bangladesh.
5 SBIT Inc.,
6 Computer Science, Louisiana State University in Shreveport.
Research Article
International Journal of Science and Research Archive, 2023, 10(02), 1438-1450.
Article DOI: 10.30574/ijsra.2023.10.2.1086
Publication history:
Received on 14 November 2023; revised on 27 December 2023; accepted on 30 December 2023
Abstract:
Adversarial attacks on artificial intelligence (AI) systems have become a growing concern in the field of cybersecurity. Such attacks are based on minor alterations in the input data that may mislead AI models and make wrong judgments, which is a serious threat to many industries, which use AI technologies, including autonomous vehicles, healthcare, and finance. The growing complexities in such attacks bring out weak points to AI systems, which poses threat to their integrity, safety and reliability. This study examines adversarial attacks and how such attacks are made and their effect on AI-based systems. The research looks at different defence strategies and their contributions towards curbing such threats. The research mentions the main issues of detecting and defending against adversarial attacks through an in-depth analysis of real-life case studies and the necessity to harness the issue with enhanced security precautions. The approach is a synthesis of case studies, simulations, and metrics of evaluation in order to understand the susceptibility of AI models. Significant details of the research include the ever-increasing mounting sophistication of attacks and the dire necessity of sturdy defense measures to secure the AI systems.
Keywords:
Adversarial attacks; Machine learning; Deep learning; Model robustness; Defense strategies; AI vulnerabilities
Full text article in PDF:
Copyright information:
Copyright © 2023 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0
