Large language models aren’t like disabled people

"There’s a tendency I’ve observed where people trying to argue that language models “understand” language to draw analogies to the experience of disabled people (especially blind and Deafblind people). These analogies are both false and dehumanizing."

there are some imperfect analogies where 'you're not wrong Walter, you're just an asshole' might apply