Human Divergent Exploration Capacity for Material Design: A Comparison with Artificial Intelligence

Applications of artificial intelligence (AI) to material design have attracted increasingattention  in  recent  years.   Although  AI-aided  material  design  holds  great  promise  forsome applications, whether it has surpassed human creativity remains uncertain.  Theaim of the current study was to compare the divergent exploration capacity of AI withthat  of  humans  on  a  material  design  task.   Human  participants  were  asked  to  find  ahigh-performance lubricant molecule under searching conditions comparable to a state-of-the-art AI system.  Results indicated that, on average, AI was able to find significantlybetter lubricant molecules.  However, the best molecule found by AI fell short of the bestmolecule found by a human participant.  Furthermore, the structural characteristics ofthe  molecules  found  by  AI  and  human  participants  differed  significantly.   These  findings suggest that a state-of-the-art AI system is capable of surpassing human divergentexploration  capacity  in  material  design,  as  in  other  fields  in  which  AI  has  advanced.Nevertheless,  our  results  also  demonstrate  that  human  intelligence  and  AI  can  playcomplementary  roles.   This  investigation  opens  up  new  possibilities  for  collaborative systems involving both AI and humans in material design.