Sign In

Communications of the ACM

ACM Opinion

Large Language Models Associate Muslims with Violence


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
Large Language Models Associate Muslims with Violence

More research is urgently needed to better debias large language models. While applications are still in relatively early stages, these tasks may be influenced by Muslimviolence bias. Dialogue generated about Muslims may include excessive violence, or news stories about Muslims may be incorrectly summarized as to suggest they have perpetrated terrorism.

Large language models, which are increasingly used in AI applications, display undesirable stereotypes, such as persistent associations between Muslims and violence.

Natural language processing (NLP) researchers must identify better ways to mitigate stereotypical bias against Muslims, as well as other social biases, that can be promoted by language models.

From Nature Machine Intelligence
View Full Article


 

No entries found