Your Trusted Partner for Affordable Photography Essentials, Backed by Quality and Care

How Google’s AI Model Swiftly Removes Watermark From Images


Generative AI has been posing new challenges for image-makers for a while now. From copyright infringement to ethical concerns, generative AI has been advancing so rapidly without any consideration of its effects on human creativity. An example is the latest development of Google’s Gemini 2.0 Flash AI model, which can now effortlessly remove watermarks from the pictures.

As reported by Tech Crunch, Gemini 2.0 Flash can clean off watermarks, including those from Getty Images and other major stock photography agencies; this has raised a controversy about legal and ethical challenges. For instance, Google’s generative AI does not only remove the watermark but also fills in the gaps left behind by the tool. This is very concerning because a watermark can’t be removed without the creator’s permission, and such removal is illegal under U.S. copyright law. The Digital Millennium Copyright Act (DMCA) forbids any alteration or removal of copyright information without consent.

However, on Reddit, people have pointed out how Google’s latest AI feature, available at Google AI Studio, is not exactly new, with features like this being around the web for a while. At the same time, they have also pointed out how people would use Photoshop to remove watermarks. The only difference is that this tech will become increasingly easy to steal someone’s work, and its accessibility (despite being in the beta stage) will further help bad actors. For instance, this will further impact stock photographers and their source of income.

Tech Crunch also got a response from Google: “Using Google’s generative AI tools to engage in copyright infringement is a violation of our terms of service. As with all experimental releases, we’re monitoring closely and listening for developer feedback.” In other words, they will see the experiment a bit more before guardrails are explicitly set in place.

So, what can you do in this mayhem to safeguard your work?

  • Better policies and guardrails are a must at this point. Claude’s AI downright refuses to remove watermarks. Similarly, Google’s AI is free to use, so it would be best for Google to have really strict rules in place if they do not wish to face repercussions and lawsuits.
  • Metadata and digital signatures have become even more important today. With your ownership in place, if you see your image somewhere used without your permission, your digital signature and metadata will help you file and win a lawsuit.
  • Educating more users about copyright challenges and the legal consequences of using an unauthorized image.

Overall, the only way out is to be vigilant and hope that your images are not used by someone who knows these tools. The challenge we face now is simply boiling down to whether you upload the image or refrain from it. Although we hope there was a much better solution for this, there isn’t. We simply have to be vigilant and do our part. If you find your work somewhere without your permission, do not hesitate to sue the person or organization. Our hope is that people will have enough understanding to avoid taking such illegal routes.



Source link

We will be happy to hear your thoughts

Leave a reply

DPSaver
Logo
Register New Account
Compare items
  • Total (0)
Compare
0
Shopping cart