Cybersecurity Strategies for Protecting Against AI-Generated Disinformation Campaigns
Saswata Dey, Sundar Tiwari, Writuraj Sarma · International Journal of Emerging Trends in Computer Science and Information Technology · 2024
AI-assisted disinformation has become a new variant of cyber threats that impacts people’s campaigns and mentalities, political processes, and economic growth. New problems, namely deep learning techniques, GANs, language models, and synthetic media, cannot be solved by simple cybersecurity approaches. The paper aspires to develop a detailed cybersecurity model that discusses the measures for countering disinformation created by AI by implementing multiple layers of protection, such as an automated detection system, real-time content check mechanism, human involvement, and intersectoral cooperation. Firstly, machine learning is used in attacking/defending mode for identifying fake or doctored media; secondly, use of blockchain for documenting media origins; thirdly, psychological profiling for counter-narrative strategies. This study provides a qualitative and quantitative assessment of the proposed framework with synthetic news articles, deepfake videos, and text generated from deep learning models. It also concerns the socio-technical aspects of disinformation, which is why the study incorporates ethical analysis, policies, and users’ awareness as its parts. The conclusion drawn concerns the effectiveness of the threat and puts into perspective that the best way to control it is by implementing technological, regulatory and behavioral approaches