Trust & Safety

AI-Generated Content Disclosure

Transparency about how Deepacal generates media and your labelling duties.

Effective date: 29 July 2026 · Applies to Deepacal, deepacal.com and all Deepacal applications.

1. Everything produced on Deepacal is synthetic media

Live face swaps, avatar videos, photo swaps, AI photo edits, character swaps, lip-sync, translations, cloned voices and scripts are generated or substantially altered by artificial intelligence models. Output is not a faithful recording of reality even when it originates from your camera.

2. Models and providers

Deepacal orchestrates third-party foundation and media models for real-time swapping, avatar synthesis, speech synthesis and cloning, image editing, translation and text generation. Providers may change as models improve. The current list is maintained in the Subprocessors & Third-Party Providers Notice.

3. Your disclosure obligations

  • Where the law requires it — including EU AI Act Article 50 transparency duties for deepfakes and various national and state statutes — you must disclose that the content is artificially generated or manipulated.
  • Where a viewer could be misled about a real person's words or actions, disclose regardless of legal requirement.
  • Do not represent Deepacal output as authentic evidence, live journalism or verified identity.

4. Provenance

Some providers embed watermarks or provenance signals in output. Removing, obscuring or spoofing them violates the Acceptable Use Policy and may violate law.

5. Accuracy

AI output can hallucinate, mistranslate, mispronounce and distort. Always review before publishing. Deepacal disclaims liability for reliance on unreviewed output.

Related legal documents

See all 31 documents in the Deepacal Legal Center.