By analyzing visual, auditory, and language inputs, this AI can predict how the human brain responds to sights and sounds with a 70x resolution increase over previous models.