Google releases new ‘open’ AI models with a focus on safety

Gemma 2 2B is a lightweight model for generating analyzing text that can run on a range of hardware, including laptops and edge devices. It’s licensed for certain research and commercial applications and can be downloaded from sources such as Google’s Vertex AI model library, the data science platform Kaggle and Google’s AI Studio toolkit.

As for ShieldGemma, it’s a collection of “safety classifiers” that attempt to detect toxicity like hate speech, harassment and sexually explicit content. Built on top of Gemma 2, ShieldGemma can be used to filter prompts to a generative model as well as content that the model generates.

Lastly, Gemma Scope allows developers to “zoom in” on specific points within a Gemma 2 model and make its inner workings more interpretable. Here’s how Google describes it in a blog post: “[Gemma Scope is made up of] specialized neural networks that help us unpack the dense, complex information processed by Gemma 2, expanding it into a form that’s easier to analyze and understand. By studying these expanded views, researchers can gain valuable insights into how Gemma 2 identifies patterns, processes information and ultimately makes predictions.”

The release of the new Gemma 2 models comes shortly after the U.S. Commerce Department endorsed open AI models in a preliminary report. Open models broaden generative AI’s availability to smaller companies, researchers, nonprofits and individual developers, the report said, while also highlighting the need for capabilities to monitor such models for potential risks.