AI content detection

ShieldGemma is a series of safe content moderation models developed by Google and built on Gemma 2, focusing on four harm categories (inappropriate content, dangerous content, hate and harassment). They are text-to-text decoder-only large language models, English only, with open weights, including models with 2B, 9B, and 27B parameter sizes. These models are designed to improve the safety of AI applications as part of a responsible generative AI toolkit.

Binoculars is an advanced AI-generated text detection tool that can be used with zero configuration without training data. Its detection idea is very simple and clear: most causal language models that only use decoders use a large number of the same data sets during pre-training, such as Common Crawl, Pile, etc. For more information about this method and its effects, see our paper Discovering LLMs with Binoculars: Zero-shot Detection of Machine-Generated Text.