AI Data Protection
Find Your Own Customer Records in AI Prompts, Without Sharing Them
- 01
Identify sensitive values
- 02
Replace with stable tokens
- 03
Restore only inside your boundary
500+
Built-in detectors for personal data and secrets
45
Country packs, checksum-validated where the format has one
9
AI providers behind one gateway, plus a no-code proxy
The problem
Pattern detectors find things that look like an email or an IBAN. They can't know that a particular name is one of your customers, or that an internal reference number belongs to a real account.
Uploading the customer list to a vendor so it can recognise those values would create the very exposure you're trying to prevent.
How Maskflare Exact Data Match works
The Maskflare command-line tool reads your CSV locally, normalises each value, and hashes it with a key specific to your organization. Only the hashes are uploaded. At request time, text in AI traffic is normalised and hashed the same way and compared, so a match is found without Maskflare ever holding your customers' values in clear text.
Benefits
Clear values stay with you
Hashing happens on your machine. The tool never uploads or prints the values in your file.
Robust to formatting
Values are normalised for case, accents, and look-alike Unicode characters, so María and MARIA both match.
Your keys
Hashes are keyed per organization, so they are useless outside your workspace.
Works with every route
Matches apply to gateway, proxy, and API traffic alike, with the action you choose per rule.
Capabilities
Field types
Text such as names, emails, and identifiers, up to 8 fields per dataset.
Capacity
Up to 10 datasets and 5 million values per organization.
Lifecycle
Upload, check status, and delete datasets from the command line, including deleting all data and its key.
Explore related capabilities
See how Maskflare applies the same masking rules to related AI and data-protection workflows.
Frequently asked questions
What is Exact Data Match?
Exact Data Match (EDM) is a data loss prevention technique that detects specific values from your own records, such as customer names or account numbers, rather than anything matching a general pattern.
Does Maskflare see our customer data?
No. The command-line tool hashes values locally with an organization-specific key and uploads only the hashes.
How do we upload a dataset?
With the Maskflare command-line tool: run maskflare edm upload with your CSV and a schema file describing the columns, using an API key created for EDM.
Is Exact Data Match on for every workspace?
It's turned on at the platform level. Ask about it in your demo.
See it on your own data
Book a 30-minute demo.
Bring your hardest prompts.
We'll show detection, masking, and restoration on your providers and data types,
and how it fits your stack.