Drawing the full picture on diverging findings: adjusting the view on the perception of art created by artificial intelligence

Abstract AI is becoming increasingly prevalent in creative fields that were thought to be exclusively human. Thus, it is non-surprising that a negative bias toward AI-generated artwork has been proclaimed. However, results are mixed. Studies that have presented AI-generated and human-created images simultaneously have detected a bias, but most studies in which participants saw either AI-generated or human-created images have not. Therefore, we propose that the bias arises foremost in a competitive situation between AI and humans. In a sample of N = 952 participants, we show that different evaluations emerge only when AI-generated and human-created pieces of art are presented simultaneously. Importantly, we demonstrate that AI art is not devalued, but rather, human art is upvalued, indicating the existence of a positive bias toward humans, rather than a negative bias. Further, we show that attitudes toward AI and empathy partially explain the different valuations of AI and human art in competitive situations.