Know what you’re looking at.

These draft terms describe how CESOERA may be used today. They have not yet been reviewed by a lawyer, and the contracting entity and governing law will be added after that review.

Estimates, not determinations

A CESOERA result is a probabilistic estimate produced by detection models. It can be wrong in both directions: AI-generated content can receive a low score and human-made content can receive a high one. Do not use a result as the sole basis for legal, employment, education, credit, housing, insurance, safety or reputational decisions about a person, and do not present a result as proof of how content was made.

Your content

Only submit content you are allowed to submit. You keep your rights in it. You allow CESOERA to process it only to produce your result and to operate and secure the service, as described in the privacy draft. We do not use it to train our models.

Acceptable use

Do not use CESOERA to break the law, to harass or target individuals, to bypass authentication, paywalls or access controls, or to probe or overload the service. We may suspend accounts or API keys that do.

Credits and plans

Scans use credits: one for an image or text, and a length-based amount for video that is shown before you confirm. A scan that fails is not charged. Credits reset monthly and do not carry over. Paid plans are shown on the pricing page but cannot be purchased yet; their terms, including refunds, will be published before they can be.

Business API

API keys are issued on approval and may be revoked for misuse. Keep them secret; you are responsible for requests made with your keys. Rate limits apply. Detections made with an organization’s keys belong to that organization.

Changes and availability