Google AI Principles

This is Google's public statement of the ethical rules it says it will follow when building and deploying AI products like Gemini, Search AI, and other tools. The most important thing to know is that Google commits not to build certain types of AI — including weapons capable of mass casualties or surveillance tools that violate human rights — but these are voluntary promises with no external enforcement mechanism. If you're concerned about how Google's AI products handle your data or affect your life, you can review Google's broader privacy controls at myaccount.google.com.

This document is Google's AI Principles framework, published at ai.google/principles/, which establishes voluntary ethical commitments and internal governance objectives for the development and deployment of Google's artificial intelligence systems — it does not constitute a binding legal agreement but functions as a public-facing policy statement. The most significant obligations it creates are internal: Google commits to designing AI to be socially beneficial, avoid creating or reinforcing unfair bias, be built and tested for safety, be accountable to people, incorporate privacy design principles, uphold high scientific excellence standards, and be made available only for uses consistent with these principles. Notably, the document explicitly lists categories of AI applications Google will not pursue, including weapons of mass destruction, surveillance tools violating international norms, and technologies whose purpose contravenes widely accepted principles of international law — an unusual and specific negative-commitment provision rarely seen in corporate AI governance documents. The framework engages the emerging EU AI Act regulatory landscape, OECD AI Principles, and aligns with voluntary commitments solicited by the US Executive Order on AI (EO 14110); while not directly invoking GDPR or CCPA, the privacy-by-design commitment has compliance implications under both. Material compliance consideration: this document is aspirational rather than contractually enforceable, meaning regulatory bodies assessing Google's actual AI practices will measure conduct against these stated principles, creating potential FTC Act Section 5 exposure if actual practices diverge materially from these public commitments.