From atomic scale to the bulk level, properties of polymers have been investigated for decades using classical modeling approaches such as quantum mechanics, molecular dynamics, coarse-grained models, finite element analysis, etc., however, these are constrained by scale integration and computational entities. Polymers are a large family of materials whose structure consists of several smaller molecules (called building blocks) and are used in many industries ranging from food packaging to aerospace, therefore, predicting their long-term performance is a grave challenge. Recent technological impacts on artificial intelligence and machine learning render them a novel method to analyze polymers differently. These innovative new approaches do not supplant current models but rather supplement them with data-driven and predictive tools which make design and discovery faster and reliable. This chapter is centered on addressing how the phase of artificial intelligence is applied to interpolating polymers at various scale levels through the simplest demonstration of how artificial intelligence could be integrated at the various scales. The case studies are also explained to enlighten how the AI-driven approaches lead to improve the prediction of important parameters such as glass transition temperature, blend miscibility, mechanical strength and degradation behavior. All in all, polymer modeling has reached an innovation era as it is becoming more predictive and reliable.