"A software developer, using the username Aloshdenny, has open-sourced their work on GitHub, claiming that all it required was 200 Gemini-generated images, signal processing, and 'way too much free time.'"
"SynthID is a near-invisible watermarking system that tags content generated by Google's AI tools, embedding itself in the pixels of images at the point of creation."
"Aloshdenny stated, 'No neural networks. No proprietary access,' emphasizing the simplicity of his method to strip the watermarks."
"Despite Aloshdenny's claims, Google asserts that the removal of SynthID watermarks is not possible without degrading the image quality."
A software developer has reverse-engineered Google DeepMind's SynthID system, claiming that AI watermarks can be removed from images. The developer, known as Aloshdenny, documented the process on GitHub, stating it required 200 Gemini-generated images and signal processing. Aloshdenny emphasized that no neural networks or proprietary access were needed. SynthID is designed to embed near-invisible watermarks in AI-generated images, making removal challenging without degrading quality. Despite the claims, Google disputes the accuracy of Aloshdenny's assertions.
Read at The Verge
Unable to calculate read time
Collection
[
|
...
]