New: Philter One, a redaction box for your office, is coming soon. Learn more →

Contact Us

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

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

Knowledge base Reference

SOC 2 and HIPAA Control Mapping for Philter AI Proxy

Map Philter AI Proxy features to SOC 2 Trust Services Criteria and HIPAA Security Rule safeguards, with guidance for your own attestations.

For Philter AI Proxy

This guide is a technical control mapping, not legal advice or a compliance opinion. SOC 2 and HIPAA obligations depend on your specific environment and scope. Work with your compliance officer, security team, and auditor to confirm how these controls apply to your program.

Prospects in regulated industries ask the same question once the Philter AI Proxy clears the technical evaluation: which of our SOC 2 and HIPAA controls does this help us satisfy, and what do we have to configure to claim it? This guide answers both. It maps the proxy’s features to the SOC 2 Trust Services Criteria and the HIPAA Security Rule safeguards, explains how the architecture affects your Business Associate Agreement chain, and lists the configuration steps that turn each mapped capability on.

The proxy is one control among many in your environment. It does not make you compliant on its own, and it does not cover the administrative and physical safeguards that are your responsibility as the operator. What it does is give your LLM traffic a defensible, auditable privacy layer, and produce the evidence reviewers expect for AI workloads.

The architecture changes the control story

Most LLM privacy questions start from the assumption that prompts reach the model provider and you have to govern that. The proxy inverts it. It sits between your application and providers like OpenAI, Anthropic, Amazon Bedrock, and Google Gemini, redacts PII and PHI from each prompt before it crosses your network boundary, and optionally scans responses on the way back. Redaction runs inside the proxy, so prompts are not sent to any other service to be redacted. The proxy is designed to keep the sensitive data it detects inside your perimeter; detection is probabilistic, so validate it against your own traffic.

That single fact does a lot of work in an audit. Data that is redacted before egress is not confidential data being transmitted to a third party, which simplifies your boundary-protection and transmission-security story. PHI that never reaches the LLM provider keeps that provider off your Business Associate Agreement chain. The proxy is self-hosted, so the vendor (Philterd) never touches your data and is not a business associate or a sub-processor. The controls below build on that foundation.

SOC 2 Trust Services Criteria

The mapping below covers the Common Criteria plus the Confidentiality, Availability, and Processing Integrity categories. Use it as a starting point for your own control matrix; your auditor will map to the specific criteria in scope for your report.

Trust Services CriteriaWhat it calls forProxy capability
CC6.1 Logical access securityRestrict access to systems and protect data with encryptionAPI keys required on every request (generated by the proxy, shown once, stored as SHA-256 hashes), mutual TLS, an HTTPS listener, and values stored at rest encrypted with AES-256-GCM
CC6.3 Least-privilege authorizationAuthorize access based on roles and needSeparate admin and proxy roles, per-key scopes (allow-lists for providers, models, and paths), and per-key policy binding
CC6.6 Boundary protectionProtect against threats from outside the system boundaryTLS to every provider with certificate verification (and a per-provider CA where needed), request hardening (body and header size caps, a connection timeout, canonical-path checks), and a concurrency limit
CC6.7 Transmission and movement of dataRestrict and protect data in transitInbound prompt redaction before egress, opt-in response scanning (streams included), TLS on every hop, and stripping of the proxy key header before forwarding so providers never see it
CC7.1 / CC7.2 Monitoring and detectionDetect, log, and respond to anomaliesA metadata-only audit log (in the proxy’s database, on standard output, and optionally in a file), a Prometheus /metrics endpoint, OpenTelemetry tracing, and /livez and /readyz health checks
A1.1 / A1.2 AvailabilityManage capacity and maintain resilienceA concurrency limit with 503 and Retry-After, health checks, readiness that waits for the detection model, and a persistent volume for the proxy’s database
C1.1 / C1.2 ConfidentialityIdentify, protect, and dispose of confidential informationPolicy-driven redaction of confidential data before egress; an audit log that records metadata only, never message content; no storage of prompt or response text. A context, when one is configured, stores its replacements encrypted and finds them by a keyed hash of each original value, never the original
PI1.x Processing integrityProcess data completely, accurately, and only as authorizedPolicy-driven redaction; response scanning with block or flag when PII is detected in a response; refusal of a request whose redaction fails, of one carrying images, audio, or files the proxy does not redact (by default), and of a response that cannot be scanned; first-match routing to the correct policy per request

HIPAA Security Rule safeguards

This maps the proxy to the technical safeguards of the HIPAA Security Rule (45 CFR Part 164, Subpart C), with the two administrative safeguards it most directly supports. The administrative and physical safeguards as a whole remain your responsibility as the covered entity or business associate operating the proxy.

Safeguard (45 CFR)RequirementProxy capability
164.312(a)(1), (a)(2)(i) Access control, unique user identificationAllow access only to authorized users, each uniquely identifiedA unique API key per client or team, optional mutual TLS, and per-key authorization scopes
164.312(b) Audit controlsRecord and examine activity in systems that handle ePHIAn audit log capturing provider, model, policy, the API key’s ID, values found by type, latency, client IP, HTTP status, and trace ID, plus the admin API’s policy and key changes
164.312(c)(1) IntegrityProtect ePHI from improper alteration or destructionResponse scanning with block or flag, and a policy-driven redaction
164.312(d) Person or entity authenticationVerify that a person or entity seeking access is who they claimAPI-key authentication and mutual TLS, independently or together
164.312(e)(1), (e)(2)(ii) Transmission security and encryptionGuard ePHI in transit and encrypt it where appropriateTLS on the listener and to every provider, mutual TLS, and certificate verification
164.308(a)(1)(ii)(B) Risk managementReduce risks and vulnerabilities to a reasonable levelRedacting PHI before it reaches the LLM provider removes an entire category of exposure and shrinks the downstream BAA chain
164.308(a)(4) Information access managementAuthorize access to ePHI appropriatelyPer-key authorization scopes and per-key policy binding (for example, a healthcare team’s key locked to a HIPAA Safe Harbor policy)

HIPAA BAA readiness

“BAA readiness” here means readiness for your program, not a signature from Philterd. For its software, Philterd does not sign a Business Associate Agreement, because self-hosted software creates no business-associate relationship to govern: Philterd never creates, receives, maintains, or transmits your PHI. The full reasoning, from the regulatory definitions up, is in Why Philterd Doesn’t Sign a BAA. Point your compliance team there when procurement raises the question.

What the proxy gives your HIPAA program is concrete and verifiable:

  • PHI handling. The proxy redacts the PHI it detects from prompts before they leave your network, according to your policy, a standard Phileas policy. When you route healthcare traffic through a HIPAA Safe Harbor policy and enable response scanning, the proxy helps keep PHI from reaching the LLM provider in either direction. Detection is probabilistic, so validate it against your own data.
  • Content non-retention. The proxy does not store the text of prompts or responses, and has no response cache. The audit log records metadata only (values found by type, never the text itself). One thing is stored: when you configure a context, to replace a value the same way across requests, the proxy keeps each replacement in its database, encrypted with your encryption key, and finds it by a keyed hash of the original value. The original value itself is not stored. Without a configured context, nothing from a request outlives it.
  • Encryption in transit. TLS protects every hop: client to proxy, and proxy to provider. Redaction runs inside the proxy, so there is no hop to a redaction service. Mutual TLS adds verified client identity for zero-trust service-to-service deployments.
  • A shorter BAA chain. Because de-identified data is no longer PHI, redacting at the edge can take downstream LLM providers off your BAA chain entirely. See Automating HIPAA Safe Harbor for that pattern.

Configuring the proxy to support your attestations

The proxy requires an API key on every request out of the box. Response scanning is off by default, and the Docker image serves HTTPS with a self-signed certificate until you configure your own. For a SOC 2 or HIPAA posture, configure these deliberately. The authoritative reference is the proxy’s documentation; this checklist points at the relevant sections.

  1. Issue keys per client or team. Create an API key for each client or team with the proxy role, and keep admin keys for the people who manage policies. A key’s secret is shown once and stored only as a hash. For zero-trust service-to-service traffic, also require client certificates (mutual TLS, see Certificates). The two combine.
  2. Apply least privilege. Limit each key to the providers, models, and paths it needs, and bind keys that handle PHI to your HIPAA policy so the client cannot choose another.
  3. Encrypt every hop. Serve your own certificate in place of the self-signed one (Certificates), and supply a per-provider CA where you use private certificates.
  4. Turn on response scanning for PHI routes. Set the route’s outbound.action to block so a response containing detected PII is refused rather than returned (Routes). Inbound redaction alone does not catch PHI the model emits.
  5. Capture the audit trail. Set proxy.audit.file, or collect standard output, and ship the audit log to your SIEM with your standard retention (Audit log). The log is metadata only, but client IPs may be personal data, so govern it accordingly.
  6. Protect the database and its key. Keep the proxy’s data volume on encrypted storage, back it up, and keep the encryption key in your secret manager; the database cannot be read without it.
  7. Protect availability. Set proxy.limits.max-concurrent-requests to bound in-flight work, and wire /livez and /readyz into your health checks. Request-rate policy belongs to your AI gateway or ingress, not the proxy.
  8. Make it observable. Scrape the Prometheus /metrics endpoint into your monitoring stack and, where your program calls for distributed tracing, enable OpenTelemetry.

What the proxy does not cover

Honest scoping matters in an audit. The proxy is a control for LLM traffic; it is not your whole compliance program. These remain your responsibility:

  • Administrative and physical safeguards. Workforce training, sanction policies, contingency planning, facility access, and the rest of the HIPAA administrative and physical safeguards sit outside any single tool.
  • The proxy’s host and database. Encryption at rest of the proxy’s data volume, management of its encryption key, OS hardening, and patching of its host are yours. Deploying Philter in a HIPAA Environment covers the same practices for Philter, and they apply here too.
  • Your application and identity provider. Authentication of end users, session management, and access to your own application are upstream of the proxy.
  • Your downstream systems. Anything you do with redacted output, and any system that still handles PHI elsewhere in your environment, is in your scope, not the proxy’s.
  • AI gateway concerns. The proxy deliberately does not do provider routing, failover, rate limiting, response caching, token quotas, or per-tenant billing. Those belong to an AI gateway such as LiteLLM, Portkey, or Kong AI Gateway, and the proxy is designed to run alongside one, redacting on the last hop before traffic leaves your network.

Used as part of a complete program, the proxy gives your LLM workloads a control set you can point an auditor at and evidence you can produce on demand. To work through how this maps to your specific environment, talk to the team.

Frequently asked questions

Does using the proxy make us SOC 2 or HIPAA compliant?
No single tool makes an organization compliant. Compliance is a property of your whole program: policies, people, and the full set of systems in scope. The proxy is a technical control that supports specific SOC 2 criteria and HIPAA safeguards for your LLM traffic. It helps you satisfy those control objectives and produce the evidence an auditor expects, but the attestation is yours, covering your environment.
Will Philterd sign a BAA so we can include the proxy in our HIPAA program?
No, and you do not need us to. The proxy is self-hosted software that runs inside your perimeter; Philterd never creates, receives, maintains, or transmits your PHI, so there is no business-associate relationship to govern. See Why Philterd Doesn't Need to Sign a BAA for the reasoning from the definitions up. The proxy's job is to help your own HIPAA program and to keep PHI from reaching the LLM provider in the first place.
Does the proxy log prompt or response content?
No. The audit log records metadata only: provider, model, policy, the API key's ID, how many values of each type were found in the request and response, latency, client IP, HTTP status, and a trace ID. The text of a prompt or response is never written to it. Note that client IP addresses may be personal data under GDPR, so treat the audit log accordingly.
Where do the operator-level configuration details live?
The proxy's own documentation is the authoritative reference for every configuration key. This guide links to the relevant sections of the Configuration and Monitoring pages as it goes.