API Data Usage Policy

Overview

Hume AI processes user data and generates annotations of expressive behavior, termed "API data." We are committed to protecting your data and maintaining transparency about how it is used.

Hume AI does not use customer-submitted data for training its own models or improving service offerings.

Custom Model API

Data submitted through the Custom Model API is exclusively used to train the specific customer's model. This data is not used to train Hume AI's general systems or any other customer's models.

Data Retention & Access

Only information uploaded via the Hume Registry endpoint (part of Custom Model API) is retained until deletion. A limited group of authorized Hume AI staff can access this data for the following purposes only:

• Support request investigation and triage
• Suspected abuse investigation and verification
• Content classifier flagging for potential platform abuse

Consumer Services

This policy applies only to API services. Non-API consumer services like Playground operate under different terms. Please refer to our data usage for consumer services FAQ for details on how data is handled in those contexts.