Whenever I read an explanation on how large language models work— not knowing the answer just writing based on the probability of what the next word should be—I’m always like, uh that sounds how like my brain works.