A website for communicating and exploring Google’s Secure AI Framework.
As AI systems go mainstream, so do the risks. Google created SAIF to help organisations build and deploy AI more securely. But the original material needed a better home.
The task was to turn a long-form policy document into a live platform that engineers, security teams and policymakers could actually use. It had to be clear, credible and built to evolve.
We focused on what would make the platform useful from day one.
Partnered with Google’s AI, security and compliance teams
Created interactive risk map, allowing readers to identify the specific risk that impact their role and internal processes quickly and easily
A simple, logical structure for browsing and referencing content
Components that could grow with new releases
Performance, accessibility and compliance from the ground up
Worked closely with Google’s AI teams to ensure site was live to support speaking events and the launch with Google’s broader AI announcements
Created a clear visual language to support the framework without over-designing it
All content was packaged up and donated to Coalition for Secure AI in September 2025 to advance industry wide AI security standards
