The Consilience
Privacy Policy
Last updated: April 2026
What The Consilience is
The Consilience is a personal knowledge engine for researchers working across traditions. It ingests your sources, extracts cross-tradition connections, and generates grounded research outputs from your own knowledge graph. Your data powers your personal research tool — not anyone else's.
What data we collect
- Text you capture — voice transcripts, typed notes, personal thoughts
- Sources you ingest — PDFs, URLs, arXiv papers, YouTube lectures, BibTeX files
- Your generated outputs — essays, synthesis, gap analyses
- Your confirmation decisions — vocabulary bridge confirmations and refinements
- Your lens settings — research domain, traditions, vocabulary terms
We do not collect: payment information, location data, browsing history outside the app, or any data not explicitly provided by you within the app.
How your data is processed — sub-processors
Every third-party service that touches your data is listed here with its specific purpose and training policy.
Anthropic (Claude API)
Purpose: AI text generation and essence extraction
Data sent: Your prompts and retrieved source chunks
Training: Anthropic does not train on API requests by default
anthropic.com/policies/privacy
OpenAI (Embeddings API)
Purpose: Converting text to vector representations for semantic search
Data sent: Text chunks from your ingested sources
Training: OpenAI does not train on API data by default
openai.com/policies/privacy-policy
Groq (Whisper API)
Purpose: Voice transcription
Data sent: Audio recordings you make in the app
Training: Groq does not train on API data
groq.com/privacy-policy
Supabase (PostgreSQL)
Purpose: Database storage for all structured data
Data sent: All source metadata, chunks, confirmations, outputs
Location: US-based servers
supabase.com/privacy
Pinecone (Vector Database)
Purpose: Semantic search over your knowledge graph
Data sent: Numerical vector representations (not raw text)
pinecone.io/privacy
Upstash (Redis Cache)
Purpose: Temporary performance caching
Data sent: Query results cached for 5 minutes, not persisted
upstash.com/privacy
The key commitment
None of the AI providers listed above use API request data to train their models by default. Your research, your sources, and your captured thinking are not used to train any AI model. Your data belongs to you.
Data export and deletion
You can export all your data at any time via the Export function in Settings. The export includes your sources, vocabulary bridges, recognition clusters, and generation history.
To request complete deletion of your data, email anuja@theconsilience.io. All data will be removed from our databases within 14 days of the request.
Contact
anuja@theconsilience.io