AI Use & Safety Policy

At CodewithChat, we leverage state-of-the-art AI systems to accelerate your development. This policy defines our commitment to data transparency, code privacy, and security safeguards.

100% Ownership

All AI-generated code is entirely yours. No royalty, no licensing restrictions, no catches.

Zero Model Training

We never use your proprietary codebase, files, prompt histories, or settings to train base models.

Enterprise Security

Direct communication with model providers uses secure SSL pipelines with strict access parameters.

1. Introduction

At CodewithChat, our mission is to empower developers of all skill levels to build production-grade, full-stack applications. To achieve this, we utilize cutting-edge generative Artificial Intelligence (AI) technologies, specifically Large Language Models (LLMs) such as Google Gemini.

This policy serves to define the scope, standards, and protective boundaries of our AI interactions. Our goal is to ensure that your work remains secure, compliant, and completely under your control, while giving you access to next-generation code synthesis capability.

2. Code Ownership & IP

We believe in complete builder sovereignty. The intellectual property rights of all code, configuration files, and designs generated by our service belong exclusively to you.

Full Ownership: You hold 100% of the copyrights for the applications synthesized.
Commercial Use: You are free to distribute, license, sell, or open-source the resulting projects under any terms of your choice.
No Royalties: CodewithChat does not charge licensing fees or royalties for code generated via the platform.

While we employ safety nets to prevent our AI from outputting copyrighted source code, developers are encouraged to carry out standard static analysis checks before deploying code in highly sensitive proprietary enterprise environments.

3. Data Privacy & Model Training

Your proprietary codebase is yours alone. We maintain strict boundaries to ensure that your data is never leaked or incorporated into public AI models:

No Base Training: We never use your custom workspace files, text prompt history, or configuration parameters to train base models.
Enterprise Boundaries: We route our AI calls via enterprise APIs that enforce zero data retention policies for training purposes.
Local Isolation: Your local code stays on your filesystem and is only referenced in context blocks temporarily to execute your queries.

4. Accuracy & Human Oversight

Generative AI is a powerful tool to assist software development, but it is not a replacement for human judgment and code reviews.

Our AI models generate code based on patterns and heuristics. In some instances, it may suggest outdated libraries, deprecated features, or code sections containing logic flaws.

It is the user's responsibility to review, compile, test, and verify all AI-generated code before launching it to production. We highly recommend running standard builds (npm run build) and automated unit tests.

5. Acceptable Use Guardrails

To maintain a safe, clean development environment, you must adhere to ethical boundaries when leveraging CodewithChat. You are strictly prohibited from utilizing the service to build:

Malicious payloads, exploits, ransomware, or keyloggers.
Systems designed to execute unauthorized vulnerability scans or DDOS attacks.
Spam delivery bots, phishing platforms, or deepfake/impersonation tools.
Software designed to infringe upon third-party copyrights or patents.

We actively monitor syntax request rates and enforce guardrails. Accounts found systematically violating safety limits may have their AI usage revoked.

6. Security Safeguards

We apply multiple defense layers to keep your session secure:

Encryption: All telemetry and prompts are encrypted in transit using TLS 1.3 and at rest using AES-256.
Sandboxed Previews: Any web preview or code compilation runs in isolated web environments, preventing malicious runtime access.
Token Sanitization: Our APIs filter out API tokens, bearer keys, and critical configuration environment variables before they are forwarded to models.

7. Policy Updates

As generative AI regulations (e.g., the EU AI Act, FTC directives) and best practices continue to evolve, we will update this policy accordingly. Any changes will be published directly on this page.

Last modified: July 22, 2026