Privacy Gateway for AI Chat

Use any AI model. The model never sees your sensitive data.

Protecto masks sensitive data such as PII and PHI before it reaches the model. Authorized users see the real values again. Tool calls still work when they need a real value.
Runs inside your infrastructure. Tool calls still work.
Bring your own model stack. No sales deck.
Trusted by regulated enterprises & agentic platforms
Inovalon
Automation Anywhere
Citrix
Bank Of Muscat Logo
Dell
Protecto Vault case study
Automation Anywhere, enterprise automation and AI platform
5,000+
Live enterprise tenants
Logged
Every reveal of original values
Isolated
Policy and retention per tenant

Use the best model without exposing sensitive data.

Many teams self-host because they cannot send sensitive data to a hosted model like GPT or Claude. Protecto changes that. Your real values stay in your environment. The model receives masked values only.

 Before the GatewayWith the Gateway
To keep sensitive data private, you had toself-host, and accept whichever models you could runUse any model.
Model choice wasa security decisiona quality decision
Frontier models wereoff limits for sensitive workavailable, because the model never sees real values
Session persistence

Keep your conversation history private.

When you send a prompt to a hosted model, the exposure is not only the prompt you just sent. The session persists on the provider’s side: every earlier message, the context built across turns, the detail a user typed three questions ago. You control none of it. Not retention, not access, not training use, not third-party sharing.

Sensitive information also accumulates differently across a conversation than it appears in one prompt. Nothing in turn four looks sensitive on its own. Turns one through four together identify a patient.

Run a self-hosted chat interface behind the Gateway and the session stays on your infrastructure. The provider only ever holds masked values, on every turn, for the life of the conversation.

Protect data without breaking your workflow.

Real timeRuns on every turn: message, uploaded file, tool call, and response.
Architecture of the Protecto Privacy Gateway for AI Chat, inside your infrastructure (on-prem, VPC, or private cloud). The user chats in LibreChat. Files are uploaded to LibreChat and to RAG file search. LibreChat sends each request to the Protecto Privacy Gateway, which validates the request, masks sensitive values and instructions, handles streaming and tool adapters, runs local OCR and stores masked embeddings. The Gateway calls the Protecto Detection Service and Privacy Vault to mask entities into placeholders, and to authorize and unmask them for allowed use. Only masked values go to the AI model providers: OpenAI, Claude, Gemini, Grok, DeepSeek, Kimi, or Ollama. Responses stream back with protected placeholders, and the Gateway unmasks them for LibreChat and the user. Tool calls from LibreChat to tools and data services carry authorized arguments, and tool results can be masked.Your infrastructure · on-prem, VPC, or private cloud3AI model providersOpenAIClaudeGeminiGrokDeepSeekKimiOllamaUserFilesUpload as text, PDF,or image2RAG / file searchChunking · vector store · retrieval6LibreChatChat UI · tool orchestration ·session historyBehind-the-scenes replayartifact1Protecto Privacy Gateway for AI ChatRequest validation& identityPrompt / instructionmaskingStreaming & tooladaptersAsync masking +progressLocal OCRMaskedembeddings5Tools & data servicesMCP · CRM · ticketing · internal APIsProtecto Detection Service + Privacy VaultEntity detection · tokenization · authorized resolution4Masked values only<PER>NAME_01</PER><ACC>ACCT_01</ACC>Streamed responsewith protectedplaceholdersTool call(with authorized args)Tool results(masked if enabled)Mask(entities → placeholders)Authorize & unmask(for allowed use)Maria LopezNAME_01NAME_01Maria LopezMariaACC-48213Architecture of the Protecto Privacy Gateway for AI Chat, inside your infrastructure (on-prem, VPC, or private cloud). The user chats in LibreChat. Files are uploaded to LibreChat and to RAG file search. LibreChat sends each request to the Protecto Privacy Gateway, which validates the request, masks sensitive values and instructions, handles streaming and tool adapters, runs local OCR and stores masked embeddings. The Gateway calls the Protecto Detection Service and Privacy Vault to mask entities into placeholders, and to authorize and unmask them for allowed use. Only masked values go to the AI model providers: OpenAI, Claude, Gemini, Grok, DeepSeek, Kimi, or Ollama. Responses stream back with protected placeholders, and the Gateway unmasks them for LibreChat and the user. Tool calls from LibreChat to tools and data services carry authorized arguments, and tool results can be masked.Your infrastructureon-prem, VPC, or private cloudUserFilesText, PDF, or image2RAG / file searchChunk, store, retrieve6LibreChatChat UI · tool orchestration · history1Protecto Privacy Gatewayfor AI ChatRequest validation& identityPrompt / instructionmaskingStreaming & tooladaptersAsync masking +progressLocal OCRMaskedembeddings5Tools & dataMCP, CRM, APIsPrivacy VaultDetect, mask, resolveMasked values only3AI model providersOpenAIClaudeGeminiGrokDeepSeekKimiOllama
Real valuesMasked valuesSolid = requestDashed = response
01

Protect the prompt

Protecto finds sensitive data and replaces it with protected values before the prompt reaches the model. (for example, John becomes WJKLS)
AI-powered detection
02

Protect files and RAG content

Protecto protects sensitive data in uploaded files and retrieved content before it enters the prompt.
Accuracy-preserving
03

Keep tool calls working

When an approved tool needs a real value, Protecto resolves it right before the tool runs. The model never sees the real value.
Zero-trust access
04

Return the right response

Protecto restores real values for the authorized user when the response comes back. (for example, WJKLS changes back to John)
Why DLP breaks here

Masking is easy. Not breaking AI workflows is hard.

Basic masking can break a tool call when the tool needs a real customer ID, account number, or name. Protecto masks the sensitive data while allowing approved tool calls to use approved real values when needed.

Tool calls

Basic masking: The tool gets a surrogate, so exact-match lookups fail and the model answers from nothing.

Protecto: Resolves supported tool arguments at the tool boundary, so approved tool calls get the real values they need.

Tokens and latency

Basic masking: Redaction markers change token usage, and rescanning at every step adds processing and latency.

Protecto: Uses consistent, format-preserving surrogates. The impact depends on value length, tokenizer, and payload size.

Time to first token

Basic masking: Scan-then-forward needs the full payload before the first model token can go out.

Protecto: Streams the model’s response back after masking.

Works with your chat and model stack.

Any chat interface

Protecto runs behind your chat interface. You can keep the interface your team already uses. LibreChat integration is available today.

Any LLM provider

Use OpenAI, Claude, Gemini, or a self-hosted model while keeping real values out of the model request. Your model choice should be about quality, not data exposure.

Chat, on purpose

Built for chat workflows. Easy to integrate, so you know where it fits before you start a pilot.
Security & compliance

Built for regulated data at enterprise scale.

Protecto combines data protection, audit logs, and ready-to-use policies for regulated data, whether deployed as SaaS, on-prem, in a private cloud, or air-gapped.

SOC 2 Type II

Audited controls

ISO 27001

Certified ISMS

HIPAA

BAA available

GDPR

EU data ready

DPDP

India privacy law

CCPA / CPRA

US privacy
Independently verified to outperform AWS Comprehend and Microsoft Presidio · audit logs on every scan, mask, and unmask.
Resources

Go deeper.

Cover Of The Guide Protect Sensitive Data Without Breaking Ai Workflows
Guide · 8 pages

Protect Sensitive Data Without Breaking AI Workflows

How to mask sensitive values for the model and still give an approved tool the real value it needs, with a worked tool-call example and a checklist for production.

Common questions

What does this add to latency per turn?

Gateway latency depends on prompt size, entity density, selected model, synchronous versus asynchronous masking, and whether tool arguments require resolution. During deployment, Protecto measures median and tail latency against your representative prompt sizes and workflows. 

A local model is a legitimate choice and plenty of good teams have made it. It is also a choice most teams made because it was the only one available, and it costs capability on every task where a frontier model is better. The Gateway does not argue against your local model. What changes is that using a hosted model like GPT, Claude, or Gemini alongside it stops being a security decision.

Yes, because the masking is context-preserving and format-preserving rather than blanket redaction. The model reasons over structurally intact, consistent values instead of blacked-out text. In a deployment at a large bank in the Middle East, semantic accuracy held up well after masking.

Protecto is built primarily for on-prem, VPC, or air-gapped deployment, so the mapping and the masked data stay inside your environment by default. Protecto maintains the authoritative token-to-value mapping in the configured Privacy Vault. The Gateway holds mappings transiently while processing a request, and your chat interface stores encoded replay data used to reconstruct masked history. Unmasking uses a per-user Protecto token. Access to the vault, chat interface artifacts, and operational logs must be restricted.

The detection and masking engine is Vault, and it is already in production at named enterprise customers. What is new is where it runs and what it does on the way back. The Gateway sits in the chat runtime path and resolves a tagged value at the tool boundary whenever it is present and the per-user Protecto token permits it, so tool calls resolve and the user sees real data while the model never held a real value.

Protecto Gateway is designed to detect and mask supported entity-based sensitive data, including common personal, health, payment, account, government, and customer identifiers, before protected chat content reaches the model provider.

The Gateway is decoupled from the chat interface, so nothing in the architecture is tied to one UI. LibreChat is the integration available today. Open WebUI and other interfaces are available on request, and the fastest path is a short call where we look at your instance.

No. Attaching raw file bytes directly to the model provider is not supported. Files are read through the chat interface and masked before their content enters the prompt, which is what keeps the control in place.

No. The Gateway does not itself authenticate the chat session. It trusts caller-supplied identity and relies on network placement, the same as the systems in front of it. It does work with AD and Okta: you keep each user’s masking and unmasking access in AD or Okta groups, and the Gateway issues a Protecto auth key with that access at login.

Protecto detects 50+ built-in entity types across personal, health, and payment data, plus custom entities you define. Detection works in English, Arabic, French, German, Spanish, Dutch, and Japanese, with more languages on request.

As SaaS, on-prem, in a private cloud, or air-gapped. Teams that self-host their chat interface usually run Protecto in the same environment, so real values stay inside it.

Protecto holds SOC 2 Type II and ISO 27001. Protecto is HIPAA compliant. When Protecto hosts the service or can access your instance, Protecto signs a BAA.

Book a demo and bring the model and chat interface you use. We run the Gateway on your representative prompts and workflows, including tool calls, and measure latency against them.

See Protecto with your own workflow.

Bring the model and workflow you want to use. We will show you how Protecto keeps sensitive values protected.
Bring your own model stack. No sales deck.
Cover of the guide Protect Sensitive Data Without Breaking AI Workflows
Technical guide

Protect sensitive data without breaking AI workflows

A practical guide to privacy controls for AI assistants, RAG, and agents. One workflow traced from request to CRM update, with what each destination may see.

By submitting you agree to Protecto’s privacy policy.

Cover of the Protecto Privacy Vault datasheet
Datasheet

Protecto Vault: sensitive data protection for AI workflows

Features that safeguard your data across agents, AI applications, RAG pipelines, MCPs, data lakes, and enterprise systems.

By submitting you agree to Protecto’s privacy policy.

New: run any AI model without exposing real values.
See how it works