This paper discusses implementing Encrypted Machine Learning models by using oblivious transfer enabling, and secure predictions of molecular quantum properties. These secure and computationally feasible models can better protect privacy in chemistry and medical sciences, such as protecting customers' medical data. Overall, the team discusses recent findings and limitations of these models.

The paper presents a new theoretical framework that allows an analytical treatment of quantum systems at strong coupling with their surroundings, bringing deep understanding of the impact of strong system-bath coupling effects on open quantum system phenomena. It describes the application of this general approach to central problems in quantum thermodynamics such as quantum thermalization, heat transport, quantum refrigeration, nanoscale thermoelectric power generators & dissipative spin chains.