AI Usage Policy

Bring Your Own Key (BYOK) Model

WarpInsights uses YOUR chosen AI provider through a bring-your-own-key (BYOK) architecture. You maintain complete control over which AI service processes your data.

Supported AI Providers

Google Gemini - Recommended model: gemini-2.5-flash-lite (fast, cost-effective)
OpenAI - Models: GPT-4, GPT-4-turbo, GPT-3.5-turbo
Anthropic Claude - Models: Claude 3.5 Sonnet, Claude 3 Opus, Claude 3 Haiku
... Others

What This Means for You

You Own the Relationship: Your website data is sent to YOUR AI provider using YOUR API credentials
You Control Privacy: Data privacy and retention are governed by YOUR AI provider's terms, not WarpInsights
You Pay Directly: AI usage costs are billed directly by your provider (not through WarpInsights credits)
You Choose Compliance: Select an AI provider that meets your specific regulatory requirements (GDPR, HIPAA, SOC 2, etc.)
You Can Switch: Change providers or revoke API access at any time through your admin dashboard

Important: You are responsible for reviewing and accepting your AI provider's terms of service, privacy policy, and data usage agreements. WarpInsights acts as a secure intermediary but does not control how your chosen AI provider handles data.

No AI Provider Yet? WarpInsights works without AI - you'll still get PageSpeed Insights, accessibility reports, and SEO analysis. AI features enhance the experience but aren't required.

Transparent Data Transmission

When you use AI features, WarpInsights sends the following data to your chosen AI provider via API:

Data We Send

PageSpeed Insights Results: Performance metrics (LCP, FCP, TBT, CLS), scores, opportunities, and diagnostics
URL Information: The website URL being analyzed
Analysis Context: Device type (mobile/desktop), analysis date, strategy used
Historical Data: Previous analysis results (only when you request comparisons in AI chat)
Your Questions: The specific questions or prompts you ask the AI
Accessibility Data: WCAG compliance results, violation details (when analyzing accessibility)

Data We DO NOT Send

Raw Website Content: Your actual HTML, CSS, JavaScript source code
User Credentials: Passwords, API keys, authentication tokens
Personal Information: Email addresses, names, or organization details beyond necessary context
Payment Information: Credit cards, billing details, subscription data
Other Organizations' Data: Only your organization's data is visible to your AI instance

API Requests Only: All AI communication happens through secure API calls. Your AI provider receives structured data (metrics and text), not access to your infrastructure or databases.

How Data Flows

1. You request AI analysis or ask a question
2. WarpInsights prepares structured data (metrics, context)
3. Data is sent via HTTPS to your AI provider's API using your key
4. AI processes the data and returns insights
5. WarpInsights displays the response to you

Review Your Provider's Policy: Different AI providers have different data retention and usage policies:

AI-Powered Data Analysis

WarpInsights uses artificial intelligence to analyze your website statistics and translate complex analytics data into natural language insights. Our AI helps you:

Understand patterns in your website performance data
Ask questions in plain English and receive instant answers
Identify trends and anomalies in your analytics
Generate recommendations based on best practices
Simplify complex metrics into actionable insights

Important: Our AI acts as an intelligent assistant that helps you make sense of your data, not as a replacement for human judgment.

Read-Only AI Analysis

No, AI does not make any changes to your environment or data. Whether you use Google Gemini, OpenAI, or Anthropic, the AI operates with strict safety principles:

Read-Only Access: AI only analyzes existing data, never modifies it
No Automated Actions: AI cannot change settings, delete files, or alter your website
No Direct Access: AI receives only structured data through API calls, not access to your systems
Analysis Only: AI provides recommendations and insights - you decide whether to act on them
Transparent Operations: Every AI interaction is logged in your WarpInsights dashboard

Think of AI as a knowledgeable consultant who reviews your performance reports and provides insights, but never has the keys to your infrastructure.

Data Privacy Note: How your data is stored and used by your AI provider is governed by THEIR privacy policy, not WarpInsights. Review your provider's terms:

Google Gemini: Typically does not use API data for model training (verify with current terms)
OpenAI: Enterprise API data is not used for training (as of their current policy)
Anthropic: Does not train on customer data submitted via API

Always verify current policies with your chosen provider.

Understanding AI Limitations

All AI models have limitations and can occasionally make mistakes. The specific limitations vary by provider, but common issues include:

Hallucinations: AI might generate plausible-sounding but incorrect information
Context Misunderstanding: May miss subtle business context or industry-specific nuances
Data Interpretation Errors: Can misinterpret complex data relationships
Outdated Information: Training data has cutoff dates - may not know the very latest algorithm changes
Overgeneralization: Might apply general rules to specific situations inappropriately
Provider Differences: Google Gemini, OpenAI, and Anthropic models have different strengths, weaknesses, and training data

Always verify: AI recommendations should be cross-checked with your domain expertise and business knowledge before implementation. No AI model is perfect, regardless of provider.

Example Limitations by Provider

Speed vs. Accuracy: Faster models (like Gemini Flash) may be less accurate than larger models (GPT-4, Claude Opus)
Context Windows: Different models handle different amounts of historical data
Specialization: Some models are better at technical analysis, others at explaining concepts simply

Experiment with different providers to find what works best for your use case.

AI as Your Team Member

WarpInsights AI should be used as another tool or team member on your team, not as a standalone solution. Here's how to integrate AI effectively:

Collaborative Approach: Use AI insights alongside human expertise
Idea Generation: Let AI suggest hypotheses and areas to investigate
Time Savings: Use AI to quickly analyze large datasets and identify patterns
Learning Tool: Learn new ways to look at your data through AI suggestions
Validation Assistant: Use AI to cross-check your own analysis and assumptions

Think of AI as having a junior analyst on your team - helpful, knowledgeable, but requiring supervision and validation of their work.

Validating Against Your Goals

AI recommendations should always be analyzed to ensure they align with your company goals and vision. Follow this validation process:

Business Context Check: Does this recommendation support our business objectives?
Resource Assessment: Do we have the resources to implement this suggestion?
Impact Analysis: What are the potential positive and negative impacts?
Stakeholder Review: Have relevant team members reviewed this recommendation?
Testing Plan: Can we test this recommendation on a small scale first?
Metrics Alignment: Does this align with our KPIs and success metrics?

Best Practice: Create a simple checklist for validating AI recommendations that includes your specific business criteria and constraints.

Our Responsible AI Principles

WarpInsights is committed to responsible AI integration and user empowerment:

User Control: You choose your AI provider and can change or disable it anytime
Transparency: We clearly explain what data is sent to AI and how it's processed
Security: All API communications use HTTPS encryption and your keys are stored securely
No Vendor Lock-In: Support for multiple AI providers prevents dependency on a single vendor
Privacy Protection: We send only necessary data to AI and never share your API keys
User Education: We provide guidance on best practices, limitations, and validation
Optional Features: AI is enhancement, not requirement - core features work without it

WarpInsights' Role

We act as a secure bridge between your analytics data and your chosen AI provider:

We prepare your data in AI-friendly formats
We transmit requests securely using your credentials
We display AI responses in an intuitive interface

We believe AI should augment human intelligence, not replace it. Our goal is to empower you with secure access to cutting-edge AI while maintaining your control over data privacy and vendor choice.

Questions or Concerns? If you have questions about data privacy, AI provider selection, or how to validate AI recommendations, our support team can help you make informed decisions.