Questions about AI-generated portraits

Does a generated portrait depict a real person?

The service asks an image model to create synthetic output rather than selecting a person from a directory. However, no generator can guarantee that an output will never resemble a real person. Review each image and regenerate if a resemblance is concerning.

Does variation guarantee anonymity?

No. Variation changes how strongly the reference guides the result, but it is not an identity-protection measurement. For privacy-sensitive work, use a high variation, compare the result with the source, obtain an independent review, and avoid publishing if the subject could still be recognized.

Are attribute sliders exact?

No. Sliders become descriptive prompt instructions. They are not biometric measurements and percentages should not be interpreted as scientific values. Results can vary because image generation is probabilistic and because some instructions compete with one another.

Can I use a result commercially?

The service permits personal and commercial use subject to its terms, the image provider's applicable terms, and local law. That permission is not a legal opinion or a guarantee that every use is risk-free. Avoid endorsement, impersonation, deceptive testimonial, and identity-document uses.

What images are retained?

Reference uploads are processed for the request and are not placed in the public gallery. Synthetic results may be kept temporarily as recent examples. Account, payment-confirmation, usage, cookie, and advertising data are handled as described in the Privacy Policy.

When should I label an image?

Label it as AI-generated or fictional whenever the surrounding context could cause a reasonable viewer to think it is a photograph of a real person. Clear disclosure is especially important in editorial, marketing, profile, educational, and documentary-style contexts.