BACKGROUND: Early recognition of steatosis (fatty liver) and fibrosis in liver health is crucial for effectively managing and preventing the possibility of liver dysfunction. Detecting steatosis helps identify individuals at risk of liver-related diseases, such as inflammation (Non-Alcoholic SteatoHepatitis, NASH) and fibrosis. Fibrosis involves the formation of scar tissue in the liver due to chronic inflammation and injury. Early recognition of fibrosis helps categorize patients based on their risk of progression to advanced liver disease. Metabolic dysfunction-Associated Steatotic Liver Disease (MASLD) leads to many outcomes, including Metabolic dysfunction-Associated Steatohepatitis (MASH), fibrosis, and cirrhosis. We aim to show that routine clinical tests supported by machine learning offer sufficient information to predict these endpoints. METHODS: The research focused on applying various operational research methods such as Linear Regression, Support Vector Machine, K-Nearest Neighbors, Decision Tree, Multi-Layer Perceptron, and Naive Bayes. RESULTS: The proposed method - FibrAIm - allows the identification of patients at risk of complications related to the conditions analyzed based on inconclusive test results. It can also identify the risk of fibrosis in those whose results appear correct. CONCLUSIONS: Given the results obtained during the trials, FibrAIm could become a valuable tool for diagnosing patients at risk of liver early steatosis and fibrosis by identifying cases based on standardized screening tests.