home.social

#watermarking — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #watermarking, aggregated by home.social.

  1. This also shows 2/3 of my concerns about #watermarking, the 3rd being that it’s applied only if you use the “right” corporate product to produce your content. Mire excited about #fingerprinting / detecting original data sources, though obviously that’s computationally way more expensive. #aiethics

    RE: https://bsky.app/profile/did:plc:dus2ftflcqjg6joeuw46sz5g/post/3mmobyu24e22f

  2. Another story of experimenting with LLMs and their guardrails. This time removing a large copyright watermark from an image.

    Will I be able to do it? Can you call me a "master jailbreaker"?

    ambience.sk/llm-stories-anothe

    BTW, you can follow my blog using RSS or using ActivityPub: @ambience

  3. Imagewmark 0.6.0 🚀

    🖼️ C++ #watermarking (much faster, lower RAM)

    🔧 Split arch: C++ for embedding, Python for extraction

    ⚙️ New test‑suite attacks: gamma, brightness‑contrast

    🐞 Fixes: .ppm/.tif support, sign‑preservation

    🔍 Blind detection

    Article:
    testbit.eu/2026/imagewmark-0.6

    #ImageWatermarking

  4. Wer es noch nicht wusste:

    Google bettet in KI-generierte Bilder, Audios und Texte steganographische Wasserzeichen ein; das ganze heißt #SynthID. Gemini kann dieses detektieren und so KI-generierten Content identifizieren.

    Darüber sind die Betrüger, die Uber Eat ans Bein pissen wollten, gestolpert.

    bsky.app/profile/caseynewton.b

    #Steganographie #Watermarking #DigitaleWasserzeichen

  5. Wer es noch nicht wusste:

    Google bettet in KI-generierte Bilder, Audios und Texte steganographische Wasserzeichen ein; das ganze heißt #SynthID. Gemini kann dieses detektieren und so KI-generierten Content identifizieren.

    Darüber sind die Betrüger, die Uber Eat ans Bein pissen wollten, gestolpert.

    bsky.app/profile/caseynewton.b

    #Steganographie #Watermarking #DigitaleWasserzeichen

  6. Wer es noch nicht wusste:

    Google bettet in KI-generierte Bilder, Audios und Texte steganographische Wasserzeichen ein; das ganze heißt #SynthID. Gemini kann dieses detektieren und so KI-generierten Content identifizieren.

    Darüber sind die Betrüger, die Uber Eat ans Bein pissen wollten, gestolpert.

    bsky.app/profile/caseynewton.b

    #Steganographie #Watermarking #DigitaleWasserzeichen

  7. Wer es noch nicht wusste:

    Google bettet in KI-generierte Bilder, Audios und Texte steganographische Wasserzeichen ein; das ganze heißt #SynthID. Gemini kann dieses detektieren und so KI-generierten Content identifizieren.

    Darüber sind die Betrüger, die Uber Eat ans Bein pissen wollten, gestolpert.

    bsky.app/profile/caseynewton.b

    #Steganographie #Watermarking #DigitaleWasserzeichen

  8. Wer es noch nicht wusste:

    Google bettet in KI-generierte Bilder, Audios und Texte steganographische Wasserzeichen ein; das ganze heißt #SynthID. Gemini kann dieses detektieren und so KI-generierten Content identifizieren.

    Darüber sind die Betrüger, die Uber Eat ans Bein pissen wollten, gestolpert.

    bsky.app/profile/caseynewton.b

    #Steganographie #Watermarking #DigitaleWasserzeichen

  9. "With the embrace of #AIvideo generation for political gain, it is likely that soon a major election will be influenced by imaginary imagery. Governments’ impulse may be to rein in video generation software like Sora, likely through mandatory #watermarking of AI content. But such controls can easily be bypassed in open-source versions of the big AI models.

    It could be far more important to ensure we have robust #publicservicejournalism."

    prospectmagazine.co.uk/ideas/t

  10. AI #watermarking was supposed to identify AI-generated images to keep the public informed. But it’s been cracked. Scientists have developed UnMark, a program that can remove AI watermarks from images. spectrum.ieee.org/ai-watermark...

  11. Engadget: Flickering lights could help fight misinformation. “The watermarking technique, called ‘noise‑coded illumination,’ hides verification data in light itself to help investigators spot doctored videos. The approach, devised by Peter Michael, Zekun Hao, Serge Belongie and assistant professor Abe Davis, was published in the June 27 issue of ACM Transactions on Graphics and will be […]

    https://rbfirehose.com/2025/07/31/engadget-flickering-lights-could-help-fight-misinformation/

  12. TechXplore: Semantic watermarks for AI image recognition can be easily manipulated. ” In a talk at the Conference on Computer Vision and Pattern Recognition (CVPR 2025) on June 15 in Nashville, Tennessee, U.S., the team revealed fundamental security flaws in the supposedly resilient watermarking techniques.”

    https://rbfirehose.com/2025/06/30/techxplore-semantic-watermarks-for-ai-image-recognition-can-be-easily-manipulated/

  13. TechXplore: Semantic watermarks for AI image recognition can be easily manipulated. ” In a talk at the Conference on Computer Vision and Pattern Recognition (CVPR 2025) on June 15 in Nashville, Tennessee, U.S., the team revealed fundamental security flaws in the supposedly resilient watermarking techniques.”

    https://rbfirehose.com/2025/06/30/techxplore-semantic-watermarks-for-ai-image-recognition-can-be-easily-manipulated/

  14. Fighting AI fakes? Google's new SynthID Detector portal helps spot their watermarked content. It's a start, but the AI world needs a unified standard for true transparency. #AI #Watermarking #DigitalTrust

  15. I've always disliked the limited #watermarking features in #LightroomClassic. I now use #Darktable, which offers a better watermark tool and supports #webp exports—both important features to me. However, for watermarking screenshots or diagrams for my blog, I turn to the #CLI tool #ImageMagick. 1/2

  16. 🚨 Breaking: OpenAI is testing watermarks for images generated via ChatGPT's free account to enhance transparency and combat misuse. 🖼️🔍 APK teardown reveals new measures are being implemented. Stay tuned! #AI #OpenAI #ChatGPT #Watermarking #ContentTransparency #TechNews #AIethics

  17. Just out of curiosity, since I'm not planning to share my recordings :anarchyblob:, does anyone knows what #watermarking tech #spotify uses these days?

    kolektiva.social/@licho/113487

  18. Just came across this insightful article on watermarking for large language models! 📝 It explores how watermarking can help distinguish between human and AI-generated text, ensuring responsible usage. A great read for anyone interested in the responsible use of AI technology!go.nature.com/3Abhjt1
    #AI #technology #technews #TextGeneration #Watermarking #EthicsInAI #ResponsibleAI #TechForGood #Innovation #NaturalLanguageProcessing #LLM #SyntheticText

  19. I'm going to the Frankfurter Buchmesse, the perfect place to meet my (e/a)bookish friends!
    Feel free to make an appointment: koalendar.com/e/meet-icontact

    #fbf24 #fbm24 #ebooks #epub #mp3 #audiobooks #ONIX #metadata #distribution #DigitalDistribution #watermarking #SocialDRM #Audiyence #BooXtream
    @frankfurterbuchmesse

  20. I'm going to the Frankfurter Buchmesse, the perfect place to meet my (e/a)bookish friends!
    Feel free to make an appointment: koalendar.com/e/meet-icontact

    #fbf24 #fbm24 #ebooks #epub #mp3 #audiobooks #ONIX #metadata #distribution #DigitalDistribution #watermarking #SocialDRM #Audiyence #BooXtream
    @frankfurterbuchmesse

  21. I'm going to the Frankfurter Buchmesse, the perfect place to meet my (e/a)bookish friends!
    Feel free to make an appointment: koalendar.com/e/meet-icontact

    #fbf24 #fbm24 #ebooks #epub #mp3 #audiobooks #ONIX #metadata #distribution #DigitalDistribution #watermarking #SocialDRM #Audiyence #BooXtream
    @frankfurterbuchmesse

  22. Tool not released because Openai believes it may lose 30% of customers.
    #OpenAI confirmed that it has developed a “highly accurate” text watermarking tool that they “continue to consider as we research alternatives”.
    However, the company said they are currently weighing the risks of such a tool, for example “our research suggests the text #watermarking method has the potential to disproportionately impact some groups,” such as non-native English speakers. #ai
    euractiv.com/section/artificia

  23. Displaying images on your website makes for an interesting problem: on one side, you want to make them publicly available; on the other, you want to protect them against undue use. The age-long method to achieve it is #watermarking.

    However, it can be much simpler to watermark them dynamically. I searched for an existing JVM library dedicated to watermarking but surprisingly found nothing. We can achieve that in a #webapp with the #Java 2D API and a simple Filter.

    blog.frankel.ch/dynamic-waterm

  24. #AI #GenerativeAI #GeneratedImages #Meta #Watermarking: "Meta is working to detect and label AI-generated images on Facebook, Instagram and Threads as the company pushes to call out “people and organisations that actively want to deceive people”.

    Photorealistic images created using Meta’s AI imaging tool are already labelled as AI, but the company’s president of global affairs, Nick Clegg, announced in a blog post on Tuesday that the company would work to begin labelling AI-generated images developed on rival services.

    Meta’s AI images already contain metadata and invisible watermarks that can tell other organisations that the image was developed by AI, and the company is developing tools to identify these types of markers when used by other companies, such as Google, OpenAI, Microsoft, Adobe, Midjourney and Shutterstock in their AI image generators, Clegg said."
    theguardian.com/technology/202

  25. #shownotes for #podcast e416 are done, and publication set for tomorrow on gamesatwork.biz and all your favorite podcast feeds! Topics this week include #selections from last week’s #Google I/O announcements, personalized #AI, #AR games, #unfoldable phones, disappearing computers, #watermarking data, and a whole lot more!

  26. for e416 are done, and publication set for tomorrow on gamesatwork.biz and all your favorite podcast feeds! Topics this week include from last week’s I/O announcements, personalized , games, phones, disappearing computers, data, and a whole lot more!

  27. #shownotes for #podcast e416 are done, and publication set for tomorrow on gamesatwork.biz and all your favorite podcast feeds! Topics this week include #selections from last week’s #Google I/O announcements, personalized #AI, #AR games, #unfoldable phones, disappearing computers, #watermarking data, and a whole lot more!

  28. #shownotes for #podcast e416 are done, and publication set for tomorrow on gamesatwork.biz and all your favorite podcast feeds! Topics this week include #selections from last week’s #Google I/O announcements, personalized #AI, #AR games, #unfoldable phones, disappearing computers, #watermarking data, and a whole lot more!

  29. #shownotes for #podcast e416 are done, and publication set for tomorrow on gamesatwork.biz and all your favorite podcast feeds! Topics this week include #selections from last week’s #Google I/O announcements, personalized #AI, #AR games, #unfoldable phones, disappearing computers, #watermarking data, and a whole lot more!