Acceptable Use Policy

Last updated 12 October 2026 · Effective 15 October 2026

You may not use the Service, or let others use it, to:

1. break the law, including any practice prohibited by Article 5 of the EU AI Act, or process content you have no right to process;

2. harm, threaten or harass individuals, groups or organisations, or incite violence;

3. promote or encourage suicide or self-harm;

4. create or spread misleading or false information, including manipulated images, audio or video presented as genuine;

5. use a real person's likeness or voice without their consent;

6. create or spread malicious code, or gain unauthorised access to systems;

7. send spam, or run phishing or other fraud;

8. create or spread content that sexualises children, or that describes or promotes the sexual exploitation of children;

9. create pornographic content;

10. create or spread terrorist, violent or extremist content, or content that promotes discrimination on the basis of protected characteristics;

11. create defamatory content;

12. develop, produce, acquire or use weapons, including chemical, biological, radiological and nuclear weapons and explosives, or use the Service for military or warfare purposes;

13. present AI-generated content as made by a human where the person receiving it is entitled to know, or fail to disclose AI-generated content where the law requires it;

14. use output to train, fine-tune or distil models, or to build services, that compete with the Inference Provider or model developer that produced it;

15. circumvent safety mechanisms, content filters, rate limits, sovereignty policies or billing at Stav or at an Inference Provider, or test the security of the Service without our written approval;

16. breach sanctions or export control rules;

17. breach the licence of a model you use, as listed at stav.ai/legal/supplier-terms, including the Llama and Gemma use policies.

Where an Inference Provider's own usage policy is stricter than this policy, it applies to requests routed to that provider. The usage policies are listed at stav.ai/legal/supplier-terms.

If you breach this policy, we may block requests, stop routing to the affected Inference Providers for you, and on serious or repeated breach suspend your workspace. Report abuse to support@stav.ai.