Phishing attacks are a growing issue and of serious concern within the Cybersecurity Domain. These attacks mostly happen from advantages taken from gaps in technology and human error, making them difficult to detect and catch. The most popular and traditional methods like Blacklists and Heuristic checks are not very affective as new phishing websites keep popping up at a faster rate to avoid detection. The presence of this gap shows the need for smarter and more flexible systems. To counter this challenge, our study brings forth an AI-driven solution that takes into use machine learning, image comparison techniques and analysing the domain info of the attacking phishing websites live when they happen. This multi-modal approach of using different methods of detection in combination makes the checks more thorough and work better with scale, making it suit for a variety of use cases. This makes the system stand out from traditional phishing detection systems due to its ability to adapt to changing nature of phishing attacks. This solution works not only for IT professionals and Security Domain experts, but also for common folks who may be at risk from this daily. We will have a thorough walkthrough of how the system is built, tech stack used for it and comparative analysis with older and more traditional methodologies for newer phishing techniques.