AI and Data Usage Policy

Introduction

NevTan Cloud is committed to providing transparent, responsible, and secure artificial intelligence services. This AI and Data Usage Policy explains how customer data, AI workloads, prompts, outputs, models, datasets, and related information are handled when customers use AI-related services offered through NevTan Cloud.

Core Principles

Our AI services are built on the following commitments:

• No training on your data: We do not use your prompts, inputs, outputs, or Customer Data to train or improve our AI models or any third-party AI models without your explicit opt-in consent.
• Data isolation: Your AI workloads, model endpoints, and inference requests are isolated from other customers.
• Transparency: We clearly disclose which AI models are available, their capabilities, and any applicable restrictions.
• Customer control: You retain ownership of your Customer Data, fine-tuned models, and AI outputs.

Inference Services

Data Processing

When you submit inference requests (prompts, images, audio, or other inputs) to hosted models on NevTan Cloud, your inputs are processed in real-time to generate the requested outputs. Inference requests are processed within NevTan Cloud's secure infrastructure.

Logging and Retention

By default, inference inputs and outputs are not persistently stored after the response is returned. Request metadata (timestamp, model, token counts, latency) is retained for billing and operational purposes. Customers may enable optional logging for debugging purposes, with configurable retention periods.

No Training Use

Your inference inputs, outputs, and prompts are never used to train, fine-tune, or improve any AI models — NevTan Cloud's own or third-party providers — without your explicit written consent.

Fine-Tuning Services

When you use NevTan Cloud's fine-tuning services, your training datasets and resulting fine-tuned model weights are treated as Customer Data. You retain ownership of fine-tuned models created using your data. Fine-tuned model weights are stored encrypted and accessible only to your account.

Bring Your Own Models

Customers may deploy their own model weights on NevTan Cloud infrastructure. Customer-provided models are treated as Customer Data. NevTan Cloud does not access, inspect, or use customer model weights except as required for technical operation of the service.

AI Output Ownership

AI-generated outputs (text, images, code, embeddings, etc.) produced using NevTan Cloud's inference services based on your inputs are owned by you, subject to the terms of the underlying model provider. NevTan Cloud makes no claim to AI outputs generated through your use of the Services.

Responsible AI Use

Customers are responsible for using NevTan Cloud's AI services in compliance with applicable laws and our Acceptable Use Policy. This includes:

• Not using AI services to generate illegal, harmful, or deceptive content
• Implementing appropriate human oversight for high-stakes AI applications
• Complying with applicable AI regulations in your jurisdiction
• Disclosing AI-generated content to end users where required by law

Third-Party AI Models

Some models available through NevTan Cloud are provided by third-party model developers (open-weight model publishers). Use of these models is subject to the respective model licenses. NevTan Cloud provides infrastructure and API access; the model weights and underlying capabilities are governed by the model developer's terms.

Security

AI workloads on NevTan Cloud benefit from the same security controls applied across the platform — encryption at rest and in transit, network isolation, access controls, and regular security assessments. See our Security Overview for details.

Contact

For AI and data policy questions, contact us at ai@nevtan.com.