Spam is an unwanted calls or SMS sent on mobile whose content may be malicious. Scammers send fake text messages or call to the user to trick people into responding to their SMS or calls. They may hack personal information, password, account number, etc. To avoid being tricked by scammers, proposed a model based on Machine Learning Algorithms. Detection of such calls is difficult as scammers evade detection through tactics such as changing numbers they call from, modifying the call script, etc. The increase of spam calls in recent years has become a major annoyance for both consumers and businesses, resulting in lost time, lower productivity, and possible privacy concerns While some smartphone applications such as true caller can detect such calls or text based on caller ID, call origination, etc., such techniques cannot easily adapt to new scams. They can detect the spam only when the phone number is registered as scam by national cybercrime. This disclosure describes the use of an AI model to detect suspicious calls. The AI model is trained on a dataset of spam/scam calls and other calls to detect spam/scam calls. With user permission, when the user receives a call from an unknown number, call content is a transcribed into text and analyzed in real time to determine if the call is likely suspicious. When such a call is detected, alerts are provided to the user by sending SMS through GSM to ensure that the user does not share sensitive information. As our project says, there are different threats or spams in real world scenario, AI model will completely analyze the call and categorize it into different form and alert the user which type of fraud it is whether it is a phishing, emotional manipulation, banking fraud, lottery scams, etc. As there were multiple threats, AI model will be trained based on the detected frauds.