A new study finds that people are more concerned about the immediate risks of artificial intelligence, like job loss, bias, and disinformation, than they are about hypothetical future threats to humanity.