AI Output Policy

CAINIX Studio uses AI-assisted analysis and generation to support manuscript review, trailer planning, scene breakdown, visual direction, prompt drafting, image generation, and related testing stages.

All AI outputs are intended to support creative decision-making, not replace it. Testers should review outputs carefully before treating them as final creative, legal, or production-ready material.

AI-generated or AI-assisted outputs may not automatically be copyrightable. Legal treatment can vary by jurisdiction, model provider, input material, and how outputs are edited or combined with human authorship.

CAINIX does not claim ownership of your uploaded manuscripts or original content. Processing inputs for requested workflow steps does not transfer ownership of your underlying materials to CAINIX.

CAINIX does not use your manuscripts or other uploaded content to train, fine-tune, or otherwise improve any AI model, ours or anyone else's, and does not add your content to any dataset. Under OpenAI's API terms, OpenAI does not train its models on API data either. Storage at the provider works as follows. Short requests are sent with OpenAI's "store" parameter set to false, so OpenAI keeps no stored copy of the response. Long, whole-manuscript steps (for example the Senior Story Director analysis, Character Census, blueprints and the screenplay) run in OpenAI's background mode, which requires OpenAI to store the response until CAINIX has read it; CAINIX then deletes that stored response, normally within minutes, and keeps retrying the deletion until OpenAI confirms it. Generated video clips are held by OpenAI's video service for a limited time so that CAINIX can download them. Separately, OpenAI may retain API inputs and outputs for up to 30 days for abuse monitoring under its standard data-usage policy; CAINIX does not currently have a zero-data-retention arrangement that would switch this off. See the Subprocessor List for the provider and more detail.

Outputs may reflect third-party model behavior, safety filters, and provider terms. CAINIX does not guarantee that outputs will be unique, accurate, or free from unintended similarity to existing works.

Some jurisdictions legally require disclosure when image, audio, or video content is artificially generated or manipulated. For example, Article 50 of the EU Artificial Intelligence Act (Regulation (EU) 2024/1689), in force since 2 August 2026, requires deployers of generative AI systems to disclose synthetic image, audio, and video content to the people who see it, in a way they can actually notice — this isn't a CAINIX policy choice, it's a legal requirement CAINIX is passing on to you. CAINIX marks the AI-generated images and video it produces accordingly (a visible label on downloaded images, a note on exported video). If you publish, share, or distribute CAINIX outputs, you are responsible for keeping that disclosure intact and for meeting any separate AI-content disclosure obligation that applies to you directly as the person distributing it, including under the laws of any jurisdiction whose audience you're distributing to.

Testers remain responsible for how they use, publish, or distribute reviewed outputs after testing.

CAINIX is operated by Uravin d.o.o., Company ID 03480836, Kosovska ulica b.b., Ulcinj, Crna Gora (Montenegro). Contact: info@uravin.me.