Research into ‘hallucinating’ generative models advances reliability of artificial intelligence

Researchers from the University of Oxford have made a significant advance toward ensuring that information produced by generative artificial intelligence (AI) is robust and reliable.