Philterd

Talk to the Team

Tell us about your stack and the privacy problems you're trying to solve. We typically respond within one business day.

Prefer email? support@philterd.ai

Prefer to skip the form? Pick a time on our calendar →
or send a message

Please do not enter PII or PHI in this form. If you need to share an example, use a sanitized one.

← All industries

Legal and E-Discovery

PII Redaction for Legal and E-Discovery

Self-hosted redaction for law firms, in-house legal teams, and e-discovery platforms. Automate the rules-of-court redactions (FRBP 9037, FRCP 5.2) and put structured exemption codes into your audit trail.

Or deploy Philter yourself →

The legal redaction problem

Legal redaction is the area where “we’ll just draw black boxes over the text” has caused more public PII leaks than any other single bad pattern. E-discovery teams spend associate hours manually finding SSNs in 50,000-page document productions. Privilege review keeps producing inconsistent outcomes because each reviewer makes slightly different calls.

The Philterd toolkit handles the structural redaction step properly (real text removal, not visible-rectangle theater) and pairs it with a human review surface for the calls that genuinely need a lawyer’s judgment.

How Philterd handles legal

FRBP 9037 / FRCP 5.2 automation

Last 4 of SSN/TIN, year-only DOB, last 4 of account numbers, initials for minors. Ready-made policy. Per-document audit log of what was detected and how it was treated.

Human-in-the-loop review

Arbiter gives reviewers a structured interface to accept, override, or annotate every automated detection with exemption codes. Audit trail attaches the reviewer’s name to every decision.

Per-document audit trail

Every detection, every modification, every reviewer action logged with timestamp and identity. The audit log is the artifact the court (or the bar) accepts.

Integrates with e-discovery platforms

Drop into pipelines that produce documents for Relativity, Logikcull, Everlaw, DISCO, or any other e-discovery platform. Pre-process before reviewer queue; reduce associate review time on mechanical redactions.

Privileged data stays in your perimeter

Attorney-client communications, work product, and client identifiers are not data you can ship to a third-party redaction API. Philter runs in your firm’s existing AWS, Azure, or GCP environment (or on-prem), with no new vendor in the data path and no new sub-processor to disclose.

Open source, auditable

Released under the permissive and business-friendly Apache license. Your security team and your IT director can audit every detection path. No black box vendor in the data flow.

Try it live

Try it out! Select one of the industries and click Redact to redact the text.

The redacted text appears here after you click Redact.

Do not enter PHI or PII.

Ready-to-use policies

Free, ready-to-use policies from the open source policy library. Download and load into your Philter instance.

Legal v1.0.0

FRCP 5.2 Federal Civil Filing Redaction

Redact federal civil filings per FRCP 5.2 — last 4 of SSN/TIN/account, year-only birthdates, minor names to initials.

FRCP 5.2federal civilcourt filingsprivacy
Legal v1.0.0

State Court Filings Baseline Redaction

A starting policy for state-court PII redaction — covers the most-common state requirements; tune for your specific jurisdiction.

state courtcourt filingsprivacyPII

Browse all redaction policies →

Recent writing on legal

All blog posts →

Where legal teams start

  1. Pick the highest-volume manual redaction task. Usually bankruptcy schedules (FRBP 9037), divorce filings, immigration documents, or e-discovery productions. The one your paralegals spend the most hours on.
  2. Deploy Philter into your firm’s infrastructure (or via the AWS Marketplace if you don’t have an existing deployment platform).
  3. Apply the relevant rules-of-court policy from the library. Tune patterns for case-management-system identifiers (case numbers, party IDs, file references).
  4. Run a representative production through it and have an associate or senior paralegal validate the output. This becomes your calibration set.
  5. Wire into the production workflow: pre-review redaction queue, attorney spot-check, then to filing. Add Arbiter when the workflow needs explicit per-reviewer attestation.

Common deployments

1. Bankruptcy practice. A firm doing volume bankruptcy work spends 10-15 hours per case on FRBP 9037 redaction across schedules, statement of financial affairs, and proof-of-claim filings. Automation cuts that to under 30 minutes of attorney review per case, with a per-document audit trail.

2. E-discovery productions. Pre-process documents through Philter before they hit the reviewer queue. The mechanical redactions (SSN, account numbers, personally identifiable financial data) are handled automatically; reviewers focus on the privilege and relevance calls that need human judgment. Architecture write-up here.

3. Court-filing preparation for in-house legal teams. Corporate legal departments preparing documents for filing or subpoena response. Run through Philter; lawyer reviews; file. Eliminates the “did the paralegal catch all the SSNs?” anxiety on every batch.

What teams need to be careful about

  • Court rules vary by jurisdiction. FRBP 9037 covers federal bankruptcy. FRCP 5.2 covers federal civil. State courts have their own (often similar but not identical) rules. The library policies are federal-baseline; verify against your specific court’s requirements.
  • Attorney responsibility doesn’t transfer to the tool. Automated redaction reduces manual work but doesn’t eliminate the lawyer’s duty to verify accuracy of filings. The audit trail helps; it isn’t a substitute for review.

Build PII redaction into your legal pipeline

Manual redaction is one of the highest-cost line items in legal operations and one of the highest-risk failure modes when it goes wrong. Talk to the engineers about whether your highest-volume workflow can be automated.

Or deploy Philter yourself →