Cyber Defense: Detecting, Predicting, and Preventing Attacks
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The capabilities of counterfeit insights (AI) have demonstrated profoundly viable in distinguishing, relieving, and overseeing cyber security dangers. In any case, later progressions have too presented modern vulnerabilities through ill-disposed AI methods, counting avoidance, harming, show reversal, AI- generated phishing, and versatile malware. This consider utilizes a mixed-methods approach, combining a study of 300 cyber security experts, investigation of real-world cases (e.g., Deep Locker), and exploratory ill-disposed control of interruption discovery and peculiarity discovery models inside AI systems. Comes about uncover that conventional security measures are viable in as it were 45% of cases, while AI based interruption discovery frameworks (IDS) make strides this to 76%, with adversarial prepared models coming to 82%. Despite these advancements, basic crevices stay, counting constrained dataset sizes, tall computational requests, show over fitting, and the nonattendance of standardized systems for AI-powered security. To address these challenges, the ponder proposes ill-disposed preparing, multi- layered AI defense designs, moral AI administration, and real - time peculiarity discovery frameworks. Moreover, it emphasizes the pressing require for quantum-resilient cyber security approaches and versatile, self-learning AI frameworks to strengthen defense components and guarantee the versatility of basic framework.
Fields of Study & Subjects
Network Security and Intrusion Detection
Adversarial Robustness in Machine Learning
Information and Cyber Security
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 26 March 2026
- Language
- EN