Press "Enter" to skip to content

Jumbled-up sentences show that AIs still don’t really understand language

Will Douglas Heaven, MIT Technology Review |

Many AIs that appear to understand language and that score better than humans on a common set of comprehension tasks don’t notice when the words in a sentence are jumbled up, which shows that they don’t really understand language at all. The problem lies in the way natural-language processing (NLP) systems are trained; it also points to a way to make them better.

Skip to content