People are using Google’s new AI model to remove watermarks from images


In social media, users have found a controversial use claim for the new Gemini AI model: delete watermarks from the pictures published with Getty photos and other well-known exchange media.

Google has expanded its access last week Twins 2.0 Flash The image generation feature of the model of the model that allows you to make and edit the image content. This is a strong abilityaccording to all accounts. However, it seems to be a few guard. Gemini 2.0 Flash will create images as imaging To describe celebrities and Copyright protected charactersand – As before – remove watermarks from existing photos.

Several x and Reddit Users noted that the twins 2.0 flash, simply delete watermarks, but try to fill any gaps created by deleting a watermark. Other AI-powered tools also do it, but the twins 2.0 flash, it looks very skilled in it – and is free to use.

To be clear, the Gemini 2.0 Flash image feature is labeled as “experimental” and “not for production use” and only available in Google-developer-looking tools Learn. The model is also a perfect watermark cleaner. Gemini 2.0 Flash fights certain semi-transparent watermarks and watermarks that canvas large parts of the pictures.

Again, some copyright holders will definitely receive an issue with the lack of restrictions on the use of twins 2.0. Models including anthropic Claude 3.7 Sonnet and Openai GPT-4O clearly refuse to remove watermarks; Claude calls remove a watermark from an image of “ethical and potential illegal”.

Eliminating a watermark without the consent of the original owner is considered illegal for the US copyright law (according to) Such as law firms( out of the rare exceptions.

Google did not respond immediately to a request for a comment sent outside the normal working hours.





Source link

Leave a Reply

Your email address will not be published. Required fields are marked *