Researchers at the University of Science and Technology of China in Hefei, China published new findings in an experiment that used AI and machine learning to determine how to create oxygen molecules on Mars. The team of researchers that included Qing Zhu, Yan Huang, Donglai Zhou, and Luyuan Zhao trained an algorithm with chemistry datasets to determine how to split water found on Mars so that future astronauts living on the planet can use the separated molecules for additional purposes such as rocket propellant and breathable air. 

The team tasked the AI model with only using minerals that were present on Mars including iron, nickel, manganese, magnesium, aluminum, and calcium. Samples were collected and studied from meteorites from the red planet found on Earth and analyzed using a special laser. The model then spent six weeks running through the various combinations of materials and temperatures until it found the optimal condition for splitting water into oxygen and hydrogen, a task that the team said would have taken 2,000 years for humans to complete.