Catalogue of Tools & Metrics for Trustworthy AI

These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.

Google Responsible AI Toolkit



The Google Responsible Generative AI Toolkit provides tools and guidance to design, build and evaluate open AI models responsibly. The toolkit supports developers in creating safe and accountable AI applications by offering resources to define rules for model behaviour and maintain transparent communication with users. It includes guidance on responsible application design, helping teams determine what type of content their application should and should not generate through system-level policies. The toolkit encourages developers to proactively identify potential risks of their applications and establish a structured approach to building safe and responsible systems.

It also provides safety alignment resources, including prompt-debugging techniques and guidance for fine-tuning and reinforcement learning from human feedback (RLHF) to align AI models with safety policies. In addition, the toolkit offers model evaluation guidance and data resources to conduct robust assessments of safety, fairness, and factuality using tools such as the LLM Comparator. Developers are also provided with safeguards, including safety classifiers that can be deployed using off-the-shelf solutions or built through step-by-step tutorials.

Through these components, the toolkit helps developers implement risk mitigation techniques and consider technical and business trade-offs when designing AI systems. It also supports the creation of transparency artefacts that communicate the responsible approach adopted for an application. Overall, the toolkit provides practical resources and documentation to apply best practices for the responsible use of open models, such as Gemma, throughout the development of generative AI applications.

About the tool

Modify this tool

Use Cases

There is no use cases for this tool yet.

Would you like to submit a use case for this tool?

If you have used this tool, we would love to know more about your experience.

Add use case
Partnership on AI

Disclaimer: The tools and metrics featured herein are solely those of the originating authors and are not vetted or endorsed by the OECD or its member countries. The Organisation cannot be held responsible for possible issues resulting from the posting of links to third parties' tools and metrics on this catalogue. More on the methodology can be found at https://oecd.ai/catalogue/faq.