AI Policy

Last updated: 24 September 2026

This AI Policy applies to everyone who uses Qloud, including your own users, and forms part of our Terms of Service. Most uses of our models are welcome. If you are unsure whether yours is, ask us at hello@qloud.sh.

1. Prohibited uses

You may not use Qloud or its Output to:

• break the law or violate anyone's rights;
• sexualize or exploit minors in any way (we report child sexual abuse material to NCMEC and law enforcement);
• plan or promote violence, terrorism, or trafficking, or seek meaningful help with biological, chemical, nuclear, or radiological weapons;
• create malware, gain unauthorized access to systems, or run phishing, spam, or denial-of-service attacks (authorized security research is allowed);
• defraud or deceive people, including impersonation, fake reviews, and election disinformation;
• create realistic images, audio, or video of real people without consent, or non-consensual intimate imagery;
• surveil, track, or identify people without authorization, including through biometric data;
• harass people or promote hatred against them, or unlawfully discriminate in decisions about jobs, housing, credit, insurance, or education;
• deliberately bypass a model's safety features; or
• use a model in breach of its license.

2. Higher-risk uses

If you use Qloud for decisions about health, legal matters, finance, employment, housing, education, public services, or critical infrastructure, you must keep a qualified human in the loop, never rely on Output as the sole basis for a decision that significantly affects someone, test for accuracy and bias, and follow the laws and professional rules that apply.

3. Transparency

Tell people when they are interacting with AI where the law requires it or where they might otherwise think they are dealing with a human, and label realistic AI-generated media of real people or events.

4. Your responsibilities

The models we serve are published by third parties, and their Output can be wrong or biased; review it before relying on it. You are responsible for your AI products and for complying with the laws that apply to them, including the EU AI Act. We do not train models on your Input or Output.

5. Reporting and enforcement

Report abuse or security issues to hello@qloud.sh with the subject "Abuse Report" or "Security Report". If we find a violation, we may warn you, restrict or suspend access, revoke keys, close your account, or report it to authorities, and we may act immediately to prevent serious harm. If you think we got it wrong, reply to our notice and we will review it.

We may update this Policy and will change the date above when we do.