The recent eras is emerging with Artificial Intelligence tools and techniques, which help to face many challenges like increasing cyber threats. The proactive cyber security strategies seem weak against the latest sophisticated cyber-attacks. Thus, the advancements in the field of AI can be advantageous as a mitigation strategy for real-time cyber risks. This paper explores security and privacy concerns and their impacts using ML models for ethical considerations. The implementation of AI in privacy-sensitive areas is discussed for its ethical impacts along with solutions for use cases of real-time applications. The ethical consideration is evaluated based on its key aspects as bias, transparency, compliance, accountability, and a few more. The research also compares the ethical concerns and compares it for available frameworks like NIST, IEEE, EU, UNESCO, ISO/IEC, etc. It is analyzed that the recent proactive strategies, as listed above are weak against most zero-day attacks. The available solutions are compared and listed with their advantages and limitations according to the attack scenario. The research concludes with mitigation strategies and suggestive actions for deploying an ethical AI framework for achieving the balance of security and privacy in evolving cyber threats.