AI Usage & Data Policy
Effective date: October 2026 · Last updated: October 2026
This policy explains how Gringo Service uses artificial intelligence, how your data is handled, and the limits of AI-generated output.
1. No Training on Your Confidential Data
We do not use client confidential data, knowledge base content or end-customer conversations to train public or third-party foundation models. Our model providers are contractually configured not to train on data we send.
2. How Agents Generate Answers
Agents use retrieval-augmented generation: responses are grounded in content you approve. Each response can be traced to its source documents.
3. Hallucination Disclaimer
Large language models can produce inaccurate output. We mitigate this with retrieval grounding, confidence thresholds, topic restrictions and testing, but cannot guarantee every answer is correct. Agents should not be relied upon for medical, legal, financial or safety-critical decisions.
4. Human Oversight
Every deployment includes human escalation paths, conversation logs and regular review by our engineers. You control which topics are automated and can pause any agent at any time.
5. AI Disclosure
Our agents identify themselves as AI assistants. We recommend clients keep this disclosure enabled.
Contact
Gerin go Technologies LLC (operating as Gringo Service), 1209 Mountain Road PL NE, STE N, Albuquerque, NM 87110, United States. Email: contact@gringoservice.com.
This document is provided for transparency and general information. It should be reviewed by qualified legal counsel before reliance in any specific jurisdiction.