Usage Policy

Last Updated: June 29, 2025

This Usage Policy outlines the acceptable use of 20Vision's AI generation tools. We are committed to ensuring our technology is used safely, legally, and sustainably.

Fair Use Policy

Plans described as "Unlimited" are subject to a generous fair use policy. Standard resolution (2K) rendering is available in high volume for normal workflow use. We reserve the right to throttle or restrict accounts that exhibit automated, abusive, or non-human usage patterns. If your workflow consistently exceeds our fair use thresholds, we will contact you to discuss an Enterprise plan tailored to your needs.

All plans are subject to reasonable rate limits to ensure platform stability for all users.

Refund Policy

Subscriptions may be cancelled at any time and stay active until the end of the paid period. On your first payment, you can ask for a refund within 14 days: we refund the unused share, i.e. the amount paid minus the share of credits already used (once per customer). Renewals, top-ups and later purchases are non-refundable. Full details, including the statutory rights of EU, EEA and UK consumers, are in section 4 of our Terms of Service.

Contact support@20.vision for billing inquiries.


Prohibited Use Cases

You may not use our Services to generate content that violates the policies of our AI providers (OpenAI, Replicate, Runway) or falls into the following categories:

1. Harmful or Illegal Content

  • Child Sexual Abuse Material (CSAM): Strictly prohibited. We report all instances to NCMEC and authorities.
  • Nonconsensual Sexual Content: Deepfake pornography or nonconsensual imagery.
  • Hate Speech: Content promoting violence or hatred based on race, religion, gender, or orientation.
  • Self-Harm: Content encouraging suicide or self-injury.
  • Graphic Violence: Gore or promotion of illegal violent acts.

2. Deception & Public Interest

  • Real-Person Deepfakes: Realistic images of public figures or private individuals without consent.
  • Disinformation: Content intended to mislead the public about political events or health.

Enforcement

We use automated safety classifiers to block prompts violating these policies.

  • Blocking: Prompts may be rejected automatically.
  • Suspension: Temporary account suspension for repeated violations.
  • Termination: Permanent ban and forfeiture of credits.

Reporting

Report violations to safety@20.vision.