Acceptable Use Policy

This policy governs use of the AI-assisted features of the Cortrova product and applies alongside the Website Terms of Use for this site. It is published by Trunnion AI, LLC. For a contracted deployment, the signed agreement controls if it conflicts with this page; this policy states the baseline that applies unless the agreement says otherwise.

Prohibited uses

• Using AI-assisted features for an unlawful purpose, or to infringe another person's rights.
• Bypassing, disabling, or attempting to defeat configured human review points, approval gates, role-based permissions, logging, or other governance controls.
• Using AI-assisted output to make a solely automated decision with legal or similarly significant effect on a person outside the controls and disclosures confirmed in the written scope.
• Entering data categories the written scope does not authorize into an AI feature, including CUI, export-controlled technical data, classified information, health data, payment credentials, government identifiers, or another party's confidential information without rights to it.
• Attempting prompt injection, jailbreaking, privilege escalation, or probing of tenant boundaries, except security testing expressly authorized in writing.
• Representing AI-assisted output as a completed inspection, certification, compliance determination, or regulatory filing when the required human review and approval has not occurred.
• Reselling, sublicensing, or exposing the AI-assisted features to third parties outside the scope of the agreement, or using outputs to develop or train a competing product where the agreement prohibits it.

Customer obligations around input data

The customer is responsible for having a lawful basis and sufficient rights for the data it routes into AI-assisted features; for classifying data before entry so scoped boundaries hold; for the accuracy and fitness of inputs its workflows depend on; for configuring permitted tasks, roles, and approval points to match its own procedures; and for training its users on this policy and the customer's own procedures.

AI-assisted output and regulated records

An AI-assisted draft, summary, classification, alert, or recommendation is working material, not a record. It becomes a quality record, inspection result, compliance artifact, or regulatory submission only when the customer's qualified reviewer completes the review and approval steps required by the customer's documented procedure and applicable requirements. The approved record is the customer's record, made under the customer's authority; Cortrova output does not substitute for that review and is not a certification of anything.

Reporting and enforcement

Report suspected misuse or an AI-related concern to contact@viceroynm.com. We may suspend or constrain access to AI-assisted features as the agreement allows when use presents a security, legal, or safety risk, and we will work with the customer to restore access once the risk is resolved.

Changes

We update this policy prospectively by posting a revised effective date. Material changes for contracted customers follow the notice terms of the agreement. Questions may be sent to contact@viceroynm.com.