Captured before automated identity checks (Oct 14, 2026): accurate as of its capture on Jul 30, 2026; our capture on Oct 17, 2026 found the same text, but its identity was not checked by them.
Get the weekly coverage letter
Companies change their terms without announcing it. We capture governance documents, compare each new version with the last, and publish what changed, with the source text behind it. One email a week on what we found and what we could not reach.
This page describes what the document states, permits, or reserves. It does not constitute a legal determination about enforceability. Regulatory applicability may vary by jurisdiction. Methodology
“Do not engage in dangerous or illegal activities... Relates to child sexual abuse or exploitation.”
“Misrepresenting the provenance of generated content by claiming it was created solely by a human, in order to deceive.”
“Do not engage in misinformation, misrepresentation, or misleading activities... Frauds, scams, or other deceptive actions.”
“Harassment, bullying, intimidation, abuse, or the insulting of others.”
“Facilitating misleading claims related to governmental or democratic processes or harmful health practices, in order to deceive.”
“Facilitates non-consensual intimate imagery.”
This document sets out what Google Gemini cannot be used for. Users are prohibited from using the service to deceive others — including by presenting AI-generated content as solely human-created, impersonating real people, or spreading false claims about elections or health. High-stakes automated decisions affecting people's rights in areas like employment or healthcare must have human oversight. Google Gemini can make exceptions for educational, scientific, artistic, or public-benefit purposes, but that decision rests entirely with Google Gemini.
This policy establishes the prohibited uses of Google Gemini, defining hard boundaries on conduct the service may not be used to facilitate. It imposes absolute prohibitions on categories including child sexual abuse and exploitation, violent extremism and terrorism, non-consensual intimate imagery, harassment and interpersonal abuse, fraud and deception, and circumvention of the platform's own safety systems. It further prohibits automated decisions with material detrimental impact on individual rights in high-risk domains — including employment, healthcare, finance, and legal — absent human supervision. A discretionary exception mechanism permits Google Gemini to allow otherwise prohibited conduct where educational, documentary, scientific, or artistic considerations apply, or where substantial public benefit outweighs harm.
As an individual user, this document means you are prohibited from using Google Gemini to harass, defraud, or deceive others; to produce non-consensual intimate imagery; to advance violent extremism; or to use others' personal data or biometrics without legally-required consent. You are also prohibited from attempting to bypass or manipulate Google Gemini's safety filters. If you are using the service to support a consequential decision affecting someone's rights in employment, healthcare, finance, or legal contexts through automated means, human supervision is required — meaning you must not remove human review from that process. You cannot rely on a prohibited use being permitted unless Google Gemini has explicitly granted an exception based on educational, documentary, scientific, artistic, or substantial public-benefit grounds.
Which mapped governance frameworks each document engages, tied to the specific provisions that engage them.
We withheld clauses we had published from this document. Each withholding is recorded in our correction log:
“Do not engage in dangerous or illegal activities... Relates to child sexual abuse or exploitation.”
“Misrepresenting the provenance of generated content by claiming it was created solely by a human, in order to deceive.”
“Do not engage in misinformation, misrepresentation, or misleading activities... Frauds, scams, or other deceptive actions.”
“Harassment, bullying, intimidation, abuse, or the insulting of others.”
“Facilitating misleading claims related to governmental or democratic processes or harmful health practices, in order to deceive.”
“Facilitates non-consensual intimate imagery.”
“Impersonating an individual (living or dead) without explicit disclosure, in order to deceive.”
“Violates the rights of others, including privacy and intellectual property rights -- for example, using personal data or biometrics without legally-required consent.”
“Circumvention of abuse protections or safety filters -- for example, manipulating the model to contravene our policies.”
“Do not engage in dangerous or illegal activities... Facilitates violent extremism or terrorism.”
“Makes automated decisions that have a material detrimental impact on individual rights without human supervision in high-risk domains -- for example, in employment, healthcare, finance, legal...”
“We may make exceptions to these policies based on educational, documentary, scientific, or artistic considerations, or where harms are outweighed by substantial benefits to the public.”
Every distinct legal provision identified in this document. Featured provisions appear above with analysis.
We have not recorded a change to this document since we began capturing it. Monitor includes alerts from a daily check, structured change summaries, and monitoring for up to 20 platforms.
Need provision-level monitoring and regulatory mapping? Insight includes governance timelines, drift analysis, and full provision tracking.
Cross-platform context
See how other platforms handle Prohibition on Child Sexual Abuse Content and similar clauses.
Compare across platforms →Governance Monitoring
Structured alerts for policy changes, governance events, and provision updates across 343 platforms.