“AI language models are trained. They're trained to produce…” — The Vergecast at 13:41
“AI language models are trained. They're trained to produce low perplexity, unsurprising sentences. Because if it's surprising, it's more likely that it's wrong. So you can build an AI detector that measures perplexity of a text and says, if it's low perplexity, it's AI. And if it's high perplexity, it's human.…”
0:00
37:48