Client Guidelines Guide
4 Core Principles
01 — Zero Training
Never use AI tools that train their models on your inputs. Always use enterprise or secured tools — LynxAI is built on zero-retention infrastructure by default.
02 — Human-in-the-Loop (HITL)
A human must review, validate, and take ownership of every AI output before it's used. Never pass AI-generated content directly to a client without human editorial review.
03 — No Shadow AI
Only use approved AI tools through LynxAI. Using personal accounts on consumer AI tools (ChatGPT free tier, Claude.ai, etc.) for client work is not permitted.
04 — Ethical Boundaries
Never use AI to deceive, discriminate, manipulate, or violate individual privacy. AI must be used to enhance our work — not to circumvent our professional and ethical responsibilities.