Transparencia de IA.Y normas de uso.

The following features are only available where the operator has configured the respective provider. AI generations consume credits; if a generation fails or is cancelled, the credits are refunded automatically.

When you start a generation, we transmit your inputs (e.g. prompt, brand profile, uploaded media or texts) to the respective AI provider. The result is stored in your workspace. No content is sent to AI providers without your action – or an agent rule you have set up. For details, including transfers to third countries, see the Privacy Policy.

The operator does not train its own AI models on your content. Processing by the AI providers is subject to their terms for use via application programming interfaces (API).

• AI outputs can be factually wrong, incomplete, outdated or biased.
• Outputs may resemble existing works, trademarks or persons; there is no guarantee of originality or freedom from third-party rights.
• AI outputs are not legal, medical, tax or financial advice.
• Review every output for content and legal compliance before publishing. You are responsible for published content.

Since 2 August 2026, the transparency obligations under Art. 50 of the AI Act (Regulation (EU) 2024/1689) apply. As a deployer of an AI system, you must in particular:

• disclose that image, audio or video content has been artificially generated or manipulated where it appreciably resembles existing persons, objects, places, entities or events and would falsely appear to be authentic (deepfakes); for evidently artistic, satirical or fictional works, a disclosure that does not hamper the display or enjoyment of the work is sufficient;
• disclose that text has been artificially generated or manipulated where it is published with the purpose of informing the public on matters of public interest – unless the text has undergone human review and a natural or legal person holds editorial responsibility.

Many platforms additionally require their own labels, such as the AI labels of TikTok, Meta (Instagram, Facebook, Threads) or YouTube. We recommend labelling AI-generated or substantially AI-edited content as a rule.

• Only clone your own voice or the voice of a person who has given you their express written consent to do so.
• Before each clone, you confirm consent and state the speaker's name. The consent is logged.
• A maximum of three voice clones per workspace per day is possible.
• Voice clones can be deleted at any time; if the speaker withdraws their consent, you must delete the clone without undue delay.
• Voice clones must never be used to impersonate another person or to deceive third parties.

In particular, it is prohibited to use flect for:

• deceptive deepfakes of real persons and depicting real persons without their consent in a degrading or misleading manner
• non-consensual intimate or sexualised imagery
• child sexual abuse material (CSAM) – such content is reported to the competent authorities
• harassment, threats, hate speech or incitement to violence
• disinformation, in particular to influence elections and democratic processes
• impersonation, fraud or phishing
• unlawful discrimination
• infringement of copyrights, trademarks, personality rights or other third-party rights
• spam, fake engagement or manipulation of platform metrics
• circumventing the safety filters of AI providers
• practices prohibited under Art. 5 of the AI Act

AI agents only act within the rules you define (e.g. topics, platforms, frequency, approval requirement). In autopilot mode, agents can schedule posts within these rules on their own. You can require approvals, pause agents at any time and edit or delete scheduled posts before they are published. You remain responsible for content published via your accounts.

AI agents do not make decisions producing legal effects concerning you or third parties within the meaning of Art. 22 GDPR.

If you notice misuse of flect, such as unauthorised use of your voice or likeness, please report it to [Email address]. We review every report and take appropriate measures, up to blocking content or accounts.

• We are transparent about which features use AI and which providers are involved.
• AI features are recognisable as such in the product.
• The operator does not train its own AI models on your content.
• We update this page when features, providers or legal requirements change.