Archives for Ethics of AI

11 Jan

How bias creeps into large language models

According to DeepMind, unmodified LMs tend to assign high probabilities to exclusionary, biased, toxic, or sensitive utterances if such language is present in the training data.

11 Jan

How bias creeps into large language models

Meeta Ramnani chatbots

According to DeepMind, unmodified LMs tend to assign high probabilities to exclusionary, biased, toxic, or sensitive utterances if such language is present in the training data.