EU AI Act Statement

1. Enforcement Timeline

The EU AI Act entered into force on 1 August 2024 and applies in phases:

2. Creofy's Role Under the EU AI Act

The EU AI Act distinguishes between providers (who develop or place AI systems on the market) and deployers (who use AI systems in the course of a professional activity). Creofy is a deployer.

We integrate third-party General-Purpose AI (GPAI) models — including image generation models (Fal.ai FLUX), video generation models (Runway), voice synthesis (ElevenLabs), and avatar video (HeyGen) — to deliver our platform's features. Each of these providers is responsible for their own model-level compliance with the EU AI Act (including GPAI transparency obligations under Article 53). Creofy is responsible for how we deploy these models and for the transparency obligations that apply at the point of content delivery.

3. AI Systems Classification

None of Creofy's features fall within the high-risk AIcategories listed in Annex III (employment, credit, law enforcement, critical infrastructure, education access, or administration of justice). No features fall within Article 5's prohibited practices (subliminal manipulation, social scoring, real-time biometric identification in public spaces).

4. Article 50 — Synthetic Media Disclosure

Article 50 of the EU AI Act requires deployers of AI systems that generate or manipulate images, audio, or video that appear realistic ("deepfakes") to ensure that the content is clearly marked as AI-generated or artificially manipulated. This obligation applies from 2 August 2026.

What this means for Creofy: All synthetic content — AI-generated images, videos, and audio — produced through our platform that depicts a realistic person or scene must carry a disclosure at the point of publication.

Measures we have implemented

• In-platform labelling:All AI-generated content in your Creofy library is marked as "AI Generated" within the platform interface. You can see the generation method, provider, and job ID for every asset.
• Publishing disclosure prompts:When you publish content to connected social platforms, Creofy reminds you to include platform-native AI disclosure labels (e.g. Instagram's "AI Info" tag, TikTok's AI-generated content label) where the platform provides this feature.
• Terms of Service obligation: Our Terms of Service require users to disclose AI-generated content when publishing, in compliance with applicable law and platform policies.
• Metadata: Where technically supported, AI-generated assets produced through Creofy include metadata indicating their synthetic origin.

Article 50 exceptions

The EU AI Act provides limited exceptions to the disclosure obligation:

• Artistic and creative works: Content that is "evidently artistic, creative, satirical, or fictional" may qualify for a reduced disclosure obligation — however, the existence of AI manipulation must still be disclosed in an appropriate manner. Creofy users relying on this exception remain responsible for ensuring their content meets the "evidently artistic" threshold.
• Editorial control: AI-generated text that has undergone human editorial review and is published under a natural person's editorial responsibility may be exempt from the text-disclosure obligation.

These exceptions are narrow. When in doubt, disclose.

5. Prohibited Practices (Article 5)

Article 5 prohibited AI practices have been enforceable since 2 February 2025. Creofy does not use or enable any of these prohibited practices:

• Subliminal or manipulative techniques to distort behaviour against users' interests
• Exploitation of vulnerabilities based on age, disability, or social/economic situation
• Social scoring systems by public or private actors
• Real-time remote biometric identification in publicly accessible spaces
• Emotion recognition in workplace or educational settings
• Biometric categorisation to infer race, political opinion, religion, or sexual orientation

6. Human Oversight & User Control

Consistent with the EU AI Act's principles of human oversight, Creofy is designed so that you remain in full control of every AI-generated output:

• No content is published to your social accounts without your explicit instruction.
• All generated content is reviewed by you before publication — nothing is auto-posted.
• You can delete any generated asset, face reference image, or voice clone at any time.
• Consent for voice cloning and face reference use is required before processing begins.

7. Data Protection & AI Act Interaction

Face reference images and voice prints processed by Creofy may qualify as biometric data under the EU AI Act and as special category data under the GDPR. We apply the following protections:

• Explicit consent is obtained before collecting or processing face or voice data.
• This data is used solely to generate your influencer's content — not to identify real persons or to train general AI models.
• Users may delete all face and voice data at any time from their account settings.
• Retention periods are set out in our Privacy Policy.

8. Third-Party Provider Compliance

As GPAI model providers under the EU AI Act, Fal.ai, Runway, ElevenLabs, HeyGen, OpenAI, and Replicate are each responsible for their own obligations under Chapter V — including technical documentation, transparency summaries, and copyright compliance for training data. We monitor our providers' compliance postures and will update our provider list if a provider fails to meet their obligations.

9. UK Alignment

The United Kingdom has not adopted the EU AI Act. The UK government is pursuing a pro-innovation, sector-led approach through existing regulators (ICO, Ofcom, FCA, CMA) rather than a single AI law. Creofy's AI transparency measures (disclosure, human oversight, consent for biometric-adjacent data) are designed to meet both the EU AI Act's requirements and the ICO's guidance on AI transparency for UK users.

10. Fines & Enforcement

The EU AI Act provides for significant fines for non-compliance:

• Prohibited practices (Article 5): Up to €35 million or 7% of total annual worldwide turnover
• Other obligations (including Article 50): Up to €15 million or 3% of total annual worldwide turnover
• Incorrect or misleading information to authorities: Up to €7.5 million or 1.5% of total annual worldwide turnover

Enforcement is carried out by national market surveillance authorities in each EU member state and coordinated by the European AI Office.

11. Updates to This Statement

We will update this statement as the EU AI Act's application phases progress (notably August 2026 and August 2027), as national implementing guidance is published, and as our platform features evolve. Material updates will be communicated via our usual notification channels.

12. Contact

For questions about our AI practices or EU AI Act compliance, contact us at support@creofy.io with the subject line "EU AI Act Enquiry".