AI Transparency
Last updated: November 24, 2025
🤖 Our Commitment to Transparency
At Levqor, we believe in being transparent about how we use artificial intelligence. This page explains when, how, and why AI is used in our platform, what data is processed, and how we protect your privacy and ensure responsible AI practices.
1. Current AI Implementation Status
Important: As of November 2025, AI features in Levqor are currently in stub/demo mode for most users. This means:
- AI features display example responses and workflows for demonstration purposes
- No actual data is sent to external AI providers unless explicitly enabled for your account
- When real AI is activated, you will be notified and can opt-in or opt-out
✅ Privacy Protection: In stub mode, your data remains entirely within Levqor's infrastructure. No external AI providers have access to your information.
2. Where AI is Used
When AI features are enabled, artificial intelligence may be used in the following areas:
AI Workflow Creator
Converts natural language descriptions into workflow configurations. Example: "Send weekly reports to my team" becomes a scheduled automation.
AI Help Panel
Provides contextual assistance and answers questions about platform features. Helps users understand errors and optimize workflows.
Debug Assistant
Analyzes error messages and suggests fixes. Helps identify common configuration mistakes and optimization opportunities.
Onboarding Guide
Provides personalized setup recommendations based on your industry and use case.
3. AI Providers & Models
When AI features are fully activated, Levqor may use the following AI providers and models:
OpenAI GPT-4o-mini
Used for natural language understanding, workflow generation, and contextual help
Provider: OpenAI LLC (United States)
Privacy Policy: openai.com/policies/privacy-policy
Note: We may add or change AI providers as technology evolves. You will be notified of significant changes.
4. What Data is Sent to AI
When you interact with AI features, the following types of data may be processed:
Data We Send:
- Workflow descriptions: Natural language descriptions you provide
- Error messages: Technical errors for debugging assistance
- Usage context: Current page, feature being used (to provide relevant help)
- Anonymized metadata: Industry, company size (for better recommendations)
Data We DO NOT Send:
- Credentials & secrets: API keys, passwords, authentication tokens
- Actual backup data: Your files, databases, or backup contents
- Payment information: Credit card numbers, billing details
- Personal identifiable information (PII): Unless you explicitly include it in a query
⚠️ Best Practice: Avoid including sensitive information (passwords, API keys, customer PII) when describing workflows or asking questions. Our AI features are designed to work with general descriptions.
5. Data Retention by AI Providers
OpenAI's data retention policy (as of our last review):
- API requests are retained for 30 days for abuse monitoring
- After 30 days, data is deleted from their systems
- Your data is not used to train OpenAI models (API data is excluded from training)
For the most current information, please review OpenAI's privacy policy directly.
6. AI Limitations & Risks
We want to be transparent about the limitations and potential risks of AI:
Hallucinations
AI may generate plausible-sounding but incorrect information. Always verify critical configurations.
Bias
AI models may reflect biases present in their training data. We actively work to mitigate this.
Context Limitations
AI has limited context about your specific business needs. Human oversight is recommended.
Security
While we implement safeguards, avoid sharing highly sensitive information in AI interactions.
7. Your Control & Rights
You have full control over AI features:
- Opt-Out: You can disable AI features entirely in your account settings
- Data Deletion: Request deletion of AI interaction history via our Data Rights page
- Review & Correct: View and correct any AI-generated configurations before saving
- Human Alternative: All AI features have manual alternatives available
8. Responsible AI Principles
Levqor is committed to responsible AI development and deployment:
- Transparency: Clear disclosure when AI is being used
- Privacy-First: Minimizing data sent to AI providers
- Human Oversight: AI assists humans, never replaces human judgment
- Accountability: We take responsibility for AI-generated content
- Fairness: Regular audits to detect and mitigate bias
- Safety: Safeguards against harmful or inappropriate outputs
9. Future AI Features
We are exploring additional AI capabilities:
- Predictive analytics for backup optimization
- Anomaly detection for unusual access patterns
- Automated workflow suggestions based on usage patterns
- Natural language query interface for data analytics
You will be notified before any new AI features are activated for your account.
📋 Legal Notice
This AI Transparency page is informational and does not override or modify our Privacy Policy or Terms of Service. For binding legal commitments, please refer to those documents.
10. Questions & Feedback
We welcome your questions and feedback about our AI practices: