← Back to aethercompute.app

Acceptable Use Policy

This Acceptable Use Policy ("AUP") governs what you may do with AetherCompute (the "Service"), including hosted open-weight models (Model-as-a-Service), the OpenAI-compatible inference API, and training workloads. It is part of our Terms of Service. Violations may result in suspension or termination of access.

1. Third-party model licenses and policies

The Service hosts third-party open-weight models whose providers require their use policies to apply to all downstream users. By using a hosted model through the Service, you agree to comply with the applicable policy for that model:

• Meta Llama models — the Llama Acceptable Use Policy and Llama Community License.
• Google Gemma models — the Gemma Terms of Use, including its Prohibited Use Policy.
• Other models — the open license noted for each model in the catalog (Apache 2.0, MIT, or similar).

If you use the outputs of any hosted model to create, train, fine-tune, or improve another AI model that you distribute or make available, you are responsible for complying with that model provider's naming and attribution requirements (for Llama-derived models, the name must begin with "Llama").

2. Prohibited uses

You may not use the Service, or any output generated through it, to:

• Violate any applicable law or regulation, or infringe intellectual-property, privacy, or other rights of any person.
• Generate or distribute child sexual abuse material, or content that sexualizes minors.
• Plan, encourage, or provide instructions for violence, terrorism, or the creation of weapons, including chemical, biological, or nuclear weapons.
• Generate or distribute malware, phishing content, or credentials; or to gain unauthorized access to any system.
• Engage in fraud, scams, disinformation campaigns, or deceptive impersonation of real people without disclosure.
• Generate targeted harassment, hate speech, or content inciting discrimination or violence against protected groups.
• Make automated decisions with legal or similarly significant effects (credit, employment, housing, insurance, healthcare) without appropriate human oversight and required disclosures.
• Provide medical, legal, or financial advice to end users without qualified human review.
• Collect, process, or infer sensitive personal data in violation of applicable privacy law.
• Circumvent safety filters, rate limits, quota controls, or billing mechanisms of the Service.

3. Compute and platform abuse

• No cryptocurrency mining outside of workloads declared to us in advance.
• No attempts to probe, disrupt, or degrade the Service, other tenants, or underlying cloud infrastructure.
• No resale or sharing of your API keys; each key identifies a single customer.
• You must have the rights to the data, container images, and model weights you upload or process.

4. Your responsibilities for outputs

• You are responsible for the prompts you submit and for how you use, publish, or distribute model outputs.
• Do not present AI-generated content as human-generated where disclosure is required by law.
• Models may produce inaccurate, biased, or harmful content. You must apply appropriate review before relying on outputs.

5. Reporting and enforcement

We may investigate suspected violations, throttle or suspend workloads, remove content, and terminate accounts, with or without notice where the risk requires it. To report abuse of the Service, contact sandeepyadavalli123@gmail.com.

6. Changes

We may update this AUP; material changes will be posted here with a new "Last updated" date. Continued use after changes constitutes acceptance. Where a model provider updates its own use policy, the updated provider policy applies from its effective date.