AI Weekly: Cutting-edge language models can produce convincing misinformation if we don’t stop them

Cutting-edge language generation models like GPT-3 are capable of producing convincing falsehoods if not properly curtailed, a new study shows.Read More