New AI model predicts record high dipole moments in unexpected molecules

Chemists may soon have one less rigorous step to worry about when searching for the right molecules to accomplish their highly specific innovation needs. Scientists have now built a new machine learning model that can predict the electric dipole moments of diatomic molecules within seconds using nothing than the atomic properties of the atoms involved. Dipole moment is the measure of charge separation between the positive and negative ions in a molecule. It is an intrinsic property of the system. In other words, it is a fingerprint of a molecule.