Two API calls. We strip PII before the LLM, restore it after. Your customers' names never touch external servers.
# Before: Risky — PII goes to external API response = openai.chat("Summarize John Smith's file at john@acme.com") # After: Safe — PII never leaves your control safe = ambientmeta.sanitize("Summarize John Smith's file at john@acme.com") # → "Summarize [PERSON_1]'s file at [EMAIL_1]" response = openai.chat(safe.text) # LLM never sees real data final = ambientmeta.rehydrate(response, safe.session_id) # → Original names and emails restored
Every company wants AI. Most can't ship it. Sound familiar?
Legal won't approve sending customer data to OpenAI
HIPAA, PCI, GDPR all say "don't send PII to third parties"
Building PII detection yourself takes 3-6 months
Your competitors are shipping while you wait
Two API calls. That's it.
Send us text, we replace PII with placeholders.
"Email John at john@acme.com"
→ "Email [PERSON_1] at [EMAIL_1]"Send the safe text to Claude, GPT-4, Gemini—whatever you want.
response = llm.complete(safe.text)Send us the response, we restore the original entities.
"I'll contact [PERSON_1]"
→ "I'll contact John"Standard entities out of the box. Plus custom patterns for your org-specific data.
John Smith, Dr. Jane Doe
john@acme.com
(555) 123-4567
123-45-6789
4532-1234-5678-9012
NYC, 123 Main St
Need to detect employee IDs or project codes? Create custom patterns →
Native integrations for popular frameworks. Model-agnostic by design.
pip install langchain-ambientmetapip install llama-index-ambientmetaDrop-in wrapper includedDrop-in wrapper includedfrom langchain_ambientmeta import PrivacyGateway gateway = PrivacyGateway(api_key="am_live_xxx") safe_llm = gateway.wrap(your_llm) # Use normally — PII handled automatically response = safe_llm.invoke("Summarize the employee file")
Start free. Pay as you grow. No credit card required.
Data sovereignty without building infrastructure.
Sensitive data stays in your control. The LLM only sees placeholders.
Self-hosted option for maximum control. Deploy in any region.
Detailed logs for every request. SOC 2 Type II in progress.
All of them. Claude, GPT-4, Gemini, Llama, Mistral—we're model-agnostic. We work at the text level, so any LLM that accepts text input works.
95%+ for standard entities like names and emails. Custom patterns can achieve 99%+. Our feedback system continuously improves accuracy.
Designed for compliance. PII never touches external APIs. Self-hosted option for maximum control. SOC 2 Type II certification in progress.
Yes. Use the /patterns endpoint to add detection for employee IDs, project codes, or any org-specific identifiers.
Less than 20ms p50 for sanitize, less than 5ms for rehydrate. Fast enough that users won't notice.
Yes. Same API, your infrastructure. Single Docker image. Starts at $500/month.
Get your API key in 30 seconds. No credit card required.
Get Free API Key