Privacy Policy for SignalPattern

Last Updated: March 18, 2026

CRITICAL PRIVACY NOTICE: SignalPattern uses multiple AI providers through our advanced model routing system. Your data will be processed by various third-party AI services including Anthropic, OpenRouter (access to 500+ models), Google, xAI, Groq (fast inference), Cerebras, Meta, Mistral AI, Cohere, DeepSeek, Qwen, Perplexity, Zhipu (GLM), and Moonshot (Kimi). Each provider has different data retention and usage policies. By using our service, you acknowledge and accept this multi-provider data processing.

1. Data We Collect

1.1 Information You Provide

  • Account Information: Email, password, name, organization
  • Project Content: All text, ideas, and information you input for analysis
  • Uploaded Files: Documents, spreadsheets, presentations you upload
  • Communication Data: Support requests, feedback, correspondence
  • Payment Information: Processed by third-party payment providers (we don't store card details)

1.2 Information We Collect Automatically

  • Usage Data: Features used, workflow patterns, interaction metrics
  • Technical Data: IP address, browser type, device information, operating system
  • Log Data: Access times, pages viewed, errors encountered
  • Performance Data: Load times, response rates, system performance

1.3 Information from Third Parties

  • OAuth Providers: Basic profile if you sign in with Google/GitHub
  • Analytics Services: Aggregated usage patterns
  • AI Providers: Processing confirmations and error reports

2. Multi-Provider AI Processing

Critical Understanding: Your data passes through multiple AI providers via our centralized model routing system. We cannot control how each provider processes your data once it leaves our systems.

2.1 Primary AI Providers

Your data may be processed by our tier-based model routing system, which includes:

  • Anthropic: Advanced language models (safety retention up to 90 days)
  • OpenRouter: Gateway to 500+ models from various providers with unified API
  • Google: Gemini models (Google Cloud data policies apply)
  • xAI: Grok models for specialized tasks (retention varies)
  • Groq: High-speed inference platform for open models
  • Cerebras: Ultra-fast inference for large language models
  • Meta: Llama open-source language models (via Groq and other providers)
  • Mistral AI: European-based language models
  • Cohere: Enterprise language models
  • DeepSeek: Specialized reasoning models
  • Qwen: Multilingual language models (Alibaba Cloud)
  • Perplexity: Search-augmented language models for real-time information retrieval
  • Zhipu (GLM): Chinese-language optimized models
  • Moonshot (Kimi): Multilingual models with native Chinese language support

Built with Llama: This service uses Meta's Llama models. Llama is licensed under the Llama Community License, Copyright Meta Platforms, Inc. All Rights Reserved.

2.2 Data Types Sent to AI Providers

The following categories of data may be transmitted to AI providers for processing:

  • Project prompts and context: Business descriptions, strategy inputs, and workflow parameters you provide
  • Intermediate analysis outputs: Results from earlier workflow stages used as context for subsequent stages
  • Search queries and grounding data: Queries derived from your project content sent to search-augmented models for real-time information retrieval
  • Structured metadata: Stage identifiers, model routing parameters, and confidence scores (no personal data)

We do NOT send your account credentials, payment information, email address, or personally identifiable information to AI providers. Only project content necessary for generating analysis is transmitted.

2.3 Data Processing Locations

Your data may be processed in:

  • United States (primary — Google Cloud Platform, Railway, Upstash, Temporal.io, Cerebras, Groq, Anthropic, xAI infrastructure)
  • European Union (Mistral AI, select OpenRouter providers)
  • United Kingdom
  • Canada
  • Singapore
  • Japan
  • China (Zhipu/GLM, Moonshot/Kimi, Alibaba Cloud/Qwen — for Chinese-language optimized models and multilingual analysis)
  • Other countries where providers operate

China-based processing: When Chinese-language models (Zhipu GLM, Moonshot Kimi, Qwen via Alibaba Cloud) are selected by our routing system, your project prompts and intermediate analysis outputs may be transmitted to servers in mainland China. These providers are used for multilingual analysis, Chinese-market research, and specialized reasoning tasks. Chinese data protection law (PIPL) applies to data processed in China. You may contact us to request that your projects not be routed through China-based providers.

2.4 Provider Data Policies

Each provider has different policies regarding:

  • Data retention periods (0-90 days typically)
  • Use for model training (some opt-out available)
  • Geographic storage locations
  • Security measures
  • Compliance certifications

3. How We Use Your Data

3.1 Service Provision

  • Process your projects through our AI workflow
  • Generate strategic insights and recommendations
  • Save and retrieve your work
  • Provide customer support

3.2 Service Improvement

  • Analyze usage patterns (anonymized)
  • Optimize workflow performance
  • Develop new features
  • Fix bugs and issues

3.3 Legal Compliance

  • Comply with legal obligations
  • Respond to legal requests
  • Enforce our terms of service
  • Protect rights and safety

4. Data Sharing and Third Parties

4.1 We Share Data With:

  • AI Providers: For processing your requests (see Section 2)
  • Infrastructure Providers: Cloud hosting, database, caching, workflow orchestration, and CDN services (see Section 4.2)
  • Service Providers: Analytics, email, support tools
  • Payment Processors: For billing and subscriptions
  • Legal Authorities: When required by law

4.2 Infrastructure and Platform Providers

Your data is stored and processed using the following infrastructure services:

  • Railway: Cloud hosting and deployment platform for our application services
  • Google Cloud Platform (GCP): Cloud SQL (PostgreSQL) for primary database storage and vector embedding infrastructure
  • Upstash: Managed Redis for caching and real-time pub/sub messaging (hosted on Google Cloud)
  • Temporal.io: Distributed workflow orchestration for reliable task execution (hosted on Google Cloud)
  • Cloudflare: DNS, CDN, and edge network services

Railway, Upstash, and Temporal.io operate on Google Cloud Platform infrastructure. Your data processed through these services is subject to both the individual provider's data handling policies and Google Cloud's infrastructure security standards.

4.3 We Do NOT:

  • Sell your personal data
  • Share content with other users without permission
  • Use your data for advertising
  • Share data beyond what's necessary for service operation

5. Data Retention

5.1 Our Retention

  • Account Data: Duration of account plus 90 days
  • Project Data: Until you delete or account termination
  • Usage Logs: 12 months
  • Support Communications: 24 months

5.2 Provider Retention

Important: AI providers have their own retention policies:

  • Some delete immediately after processing
  • Others retain for 30-90 days
  • Some may retain longer if not opted out
  • We cannot force deletion from provider systems

6. Data Security

6.1 Our Security Measures

  • Encryption in transit (TLS/SSL)
  • Encryption at rest for stored data
  • Access controls and authentication
  • Regular security audits
  • Incident response procedures

6.2 Security Limitations

  • No system is 100% secure
  • We cannot control provider security
  • Internet transmission has inherent risks
  • You're responsible for account security

7. Your Rights

7.1 You Have the Right To:

  • Access: Request copy of your data
  • Correction: Update inaccurate data
  • Deletion: Request data deletion (with limitations)
  • Portability: Export your data
  • Objection: Object to certain processing
  • Restriction: Limit how we use your data

7.2 Limitations

  • Cannot delete data already processed by AI providers
  • Some data required for legal compliance
  • Deletion may affect service functionality

8. International Data Transfers

  • Data processed globally by AI providers
  • May not have same protections as your country
  • Standard contractual clauses where applicable
  • You consent to international transfer by using service

9. Cookies and Tracking

9.1 We Use:

  • Essential Cookies: For authentication and security
  • Functional Cookies: To remember preferences
  • Google Analytics: We use Google Analytics to understand website usage, improve our services, and analyze user behavior. Google Analytics collects anonymized data including page views, session duration, bounce rates, and general geographic location (city/country level). This data helps us optimize user experience and service performance. You can opt-out of Google Analytics tracking by using the Google Analytics Opt-out Browser Add-on

9.2 We Don't Use:

  • Advertising cookies
  • Cross-site tracking for marketing
  • Social media pixels
  • Third-party advertising networks

9.3 Google Analytics Specifics:

  • Data Collected: Page URLs, referrer URLs, browser type, device type, IP address (anonymized)
  • Purpose: Website optimization, usage analysis, performance monitoring
  • Data Sharing: Analytics data is processed by Google under their privacy policy
  • Retention: Analytics data retained for 26 months
  • Your Control: Use browser settings to block cookies or install the Google Analytics opt-out extension

10. Children's Privacy

  • Service not intended for under 18
  • We don't knowingly collect children's data
  • Contact us if you believe we have children's data

11. Changes to This Policy

  • We may update this policy periodically
  • Material changes notified via email or service
  • Continued use after changes means acceptance
  • Review periodically for updates

12. Contact Information

Privacy Officer
Email: [email protected]

Data Protection Inquiries:
Email: [email protected]

General Support:
Email: [email protected]

Remember: By using SignalPattern, you accept that your data will be processed by multiple AI providers through our centralized model routing system, each with varying privacy practices. Our 12-tier routing system (Utility, Basic, Intermediate, Strategic, Advanced, Search, Vision, Math, Creative, Code, Reasoning, and Cultural tiers) routes requests across 15+ providers to optimize for performance, specialization, and privacy considerations. Always review what data you input and avoid sharing sensitive personal information.