The rapid spread of fake news on digital platforms poses a significant threat to public trust and informed decision-making, highlighting the limitations of traditional binary classification approaches that label information strictly as true or false. This research proposes an innovative fake news detection framework that explicitly models uncertainty through probabilistic reasoning techniques. By capturing the inherent ambiguity and overlap present in real-world news content, the proposed approach enables a more nuanced and realistic representation of information credibility. The model leverages probabilistic features derived from linguistic patterns, contextual cues, and source reliability to estimate confidence levels rather than absolute classifications. This uncertainty-aware framework improves robustness when handling ambiguous or incomplete information, enhancing detection accuracy and reducing misclassification risks.