AI Usage Policy — Contextlab

Last updated: July 2026

Responsible AI

We design workflows that minimize hallucination and encourage human verification.

All AI output may contain errors. Treat it as a draft, not a source of truth.

Important decisions must be validated by a human. We provide sources, rules, and diffs to help review.

You are responsible for the content you generate and the context you upload.

You choose the provider that fits. We don't lock you into a single model.

Bring Your Own Key gives you full control over cost, quotas, and provider retention.

Hash-based response cache to reduce cost. Cache never contains re-identifiable PII.

Workspace data is retained while your account is active. Export and deletion are available at any time.

Models may change due to provider policies. We announce material changes.

Every plan has quotas — see the Pricing page for details.

We document how AI is used in every feature.

AI output on Contextlab does not constitute legal, medical, or financial advice.