LangSmith Launches LLM Gateway to Redact PII in AI Requests
LangChain · x · 2026-08-10
LangChain announced new Data Protection Controls for its LangSmith LLM Gateway, now in public beta for Enterprise users.
The feature automatically detects, redacts, and replaces PII and secrets (like API keys) in requests before they reach model providers. It restores original values in the model provider's response, ensuring redaction doesn't impact agent functionality. Sensitive data is also redacted from the traces recorded in LangSmith.
More from coding & agent
- Human Behavior Launches Autonomous Analytics Agent with $5M Funding — thisguyknowsai · 2026-08-11
- The AI Coding Sweet Spot: Domain Experts Who Can Wield Agents — brandon_galang · 2026-08-11
- New Programmable Video Editor Built for AI Agents and Motion — LinusEkenstam · 2026-08-11
- New Research: Programmatic Tool Calling Beats Native JSON in Accuracy — dair_ai · 2026-08-11
- Mastra Observability Launches Automated PII Redaction for Agent Traces — massimosgrelli · 2026-08-11
- MiniMax H3 Workflow Bug: Prompt Leakage and GPU Fans Spiking to 100% — gerentedesuruba · 2026-08-11