With the advancement of generative AI (GenAI) models, new cybersecurity risks have surfaced of which including threats such as GenAI-driven phishing and deepfakes. Exacerbating the issue herein is that existing security methods and frameworks largely struggle to address these evolving attack vectors, thus making the case for adaptive mitigation strategies. This study analyzes emerging GenAI cyber threats, real-world cases, and defense mechanisms, using thematic analysis to categorize risks and solutions. It highlights the need for further studies in developing advanced threat detection methods, fortifying AI governance, and interdisciplinary collaboration, and focusing on advancing explainable AI, adversarial training, and regulatory enforcement to balance security and innovation. This research aims to strengthen our defenses against GenAI cyber threats and safeguard our digital landscape.