![](https://i.kinja-img.com/image/upload/c_fit,q_80,w_636/6574b70918ea07fece41750128a43836.png)
A new paper found that large language models from OpenAI, Meta, and Google, including multiple versions of ChatGPT, can be covertly racist against African Americans when analyzing a critical part of their identity: how they speak.
A new paper found that large language models from OpenAI, Meta, and Google, including multiple versions of ChatGPT, can be covertly racist against African Americans when analyzing a critical part of their identity: how they speak.