Model Cards / Google DeepMind

AI Principles

constitution28 words·1 min read·Apr 8, 2026·Source
Summary

AI Principles

A 132-word brief of a 28-word document. Published by Google DeepMind. Version dated Apr 8, 2026.
01

What this is

This document, titled "AI Principles," is published by Google DeepMind with a version date of 2026-04-08. The source text consists solely of a title card image for a "Responsible AI Progress Report" dated February 2026 and a link to an external PDF. No substantive content about the principles themselves is present in the provided text.

02

Capabilities

Not disclosed in this document.

03

Evaluation methodology

Not disclosed in this document.

04

Safety testing

Not disclosed in this document.

05

Mitigations

Not disclosed in this document.

06

Deployment and access

The document references an external PDF at a Google-hosted URL described as a "Responsible AI Progress Report" published in February 2026. No further access or licensing terms are disclosed in the provided text.

07

Limitations

Not disclosed in this document.

08

What's new

The card does not include version history, changelog entries, or comparison to prior versions of Google DeepMind's AI Principles.

Generated by Claude sonnet from the cleaned source on Apr 23, 2026. Passages in double quotes are verbatim from the source; other text is neutral paraphrase. For citation, use the original: original document · source SHA b1d105d16118.