ExtractFoxWhy we don't store your documents
A short note on how ExtractFox processes files in-flight, what we keep, and what we throw away — and why that's a deliberate product decision.
Most documents people send us are sensitive. Invoices have supplier prices. Bank statements have account numbers. Passports have everything. The least surprising thing we can do with that data is not keep it.
How processing actually works
A file comes in, gets sent to the multimodal model, the structured output comes back, and the source file is dropped. We keep the result long enough for you to download it, then it's gone too. There's no archive, no training set, no shadow copy waiting to leak.
What we do keep
- Account-level metadata: who you are, what plan you're on.
- Usage counters: how many extractions you've run, for billing.
- Error reports: anonymized failure traces so we can fix what's broken.
Why this is a product decision, not a compliance one
Storage costs nothing. Storage that someone might breach costs everything. The smaller the surface area, the smaller the conversation when a customer's IT team asks where their data went.
If you're evaluating ExtractFox for a regulated workflow, this is worth a longer conversation. Email hello@extractfox.com.