As members of the public increasingly turn to AI chatbots to understand their world, even subtle latent biases in the underlying models could affect public understanding of the present—and past.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results