#synthid

[aktionfsa-blog-blog]

KI-generiertes mit Wasserzeichen kennzeichnen

SynthID soll Open Source werden

Das ist eine Forderung, die das KI-Gesetz der Europäischen Union nach dem Ende der Übergangsfrist am 2. August 2026 von Anbietern verlangt. Diese Kennzeichnung darf auch nicht irgendwo versteckt angebracht werden, sondern soll dann maschinenlesbar sein. Über den dafür notwendigen Standard muss allerdings noch entschieden werden.

Dass dies unbedingt notwendig ist, hatte  eine länderübergreifende Studie des CISPA, des Helmholtz-Zentrum für Informationssicherheit in Zusammenarbeit, mit mehreren deutschen Universitäten im Mai 2024 festgestellt.

Zeit.de zitiert aus der Studie: "KI-generierte Bilder, Texte und Audiodateien sind so überzeugend, dass Menschen diese nicht mehr von menschengemachten Inhalten unterscheiden können." Mehr als 3000 Menschen waren nicht in der Lage Inhalte der KI-Programme ChatGPT oder Midjourney von "echten" Webinhalten zu unterscheiden.

Die Google-Tochter DeepMind ermöglicht bereits seit August 2023 die Verwendung eines Wasserzeichens in KI-generierten Bildern mit einem Verfahren namens SynthID. Auch für die Texterstellung mit dem Sprachmodell Gemini ist SynthID inzwischen verfügbar. Google hat nun angekündigt SynthID ab sofort kostenlos und als Open-Source-Anwendung für Entwickler und Unternehmen anzubieten.

Damit ist wieder einer der großen Internetkonzerne schneller mit der "Standardsetzung" als die Politik ...

Mehr dazu bei https://www.zeit.de/digital/internet/2024-10/ki-generierte-medien-fotos-texte-identifikation-wasserzeichen

Kategorie[21]: Unsere Themen in der Presse Short-Link dieser Seite: a-fsa.de/d/3Dv Link zu dieser Seite: https://www.aktion-freiheitstattangst.org/de/articles/8946-20241027-ki-generiertes-mit-wasserzeichen-kennzeichnen.html

[prcg]

Google lanza tecnología para marcar con agua el texto generado por IA

Google está haciendo que SynthID Text, su tecnología que permite a los desarrolladores marcar con agua y detectar texto generado por modelos generativos de IA, esté disponible de forma generalizada. SynthID Text se puede descargar desde la plataforma AI abrazando la cara y Google actualizado Kit de herramientas de GenAI responsable. “Hoy, estamos abriendo el código fuente de nuestra herramienta…

[megataskdubai]

Google New Tool To Detect AI-Generated Images

Contact Now ⤵️

📲 +971-54-587-1570 📩 info@megataskweb.com

[mpsiteweb]

Google Mengenalkan SynthID

Di era di mana kecerdasan buatan (AI) terus mengembangkan kapabilitasnya, Google telah mengambil langkah inovatif dengan meluncurkan SynthID. Produk terbaru ini adalah terobosan dalam dunia pengenalan gambar yang dihasilkan oleh AI, membawa pengalaman identifikasi gambar ke tingkat yang lebih tinggi dan akurat. Mari kita telusuri lebih jauh tentang apa itu SynthID dan mengapa hal ini menjadi berita yang begitu menarik dalam dunia teknologi.

Menggali Konsep SynthID: Identifikasi Gambar Masa Depan

SynthID adalah sistem pengenalan gambar yang menerapkan pendekatan revolusioner berbasis kecerdasan buatan. Produk ini dirancang untuk mengidentifikasi gambar yang dihasilkan oleh AI, tidak hanya dengan keakuratan yang lebih tinggi, tetapi juga dengan kecepatan dan efisiensi yang luar biasa. Teknologi ini mampu mengatasi tantangan dalam mengenali gambar yang kompleks dan terkadang samar, seperti lukisan yang dibuat oleh AI atau gambar sintetis.

Mengapa SynthID Penting?

Kontrol dan Keamanan: Dalam lingkungan di mana manipulasi gambar semakin canggih, SynthID membantu mengidentifikasi gambar yang telah dihasilkan oleh AI, sehingga memitigasi risiko pemalsuan atau manipulasi yang tidak sah. Validasi Konten Visual: Dalam konteks media dan platform berbagi, SynthID dapat membantu mengonfirmasi keaslian gambar dan menghindari penyebaran informasi yang salah atau manipulasi gambar yang merugikan. Pengembangan AI Lebih Lanjut: Dengan mampu mengenali gambar yang dihasilkan oleh AI, SynthID berpotensi memberikan umpan balik berharga kepada para pengembang AI untuk menghasilkan gambar yang lebih realistis dan valid. Bagaimana SynthID Bekerja?

SynthID menggabungkan teknik pengenalan gambar yang kuat dengan pembelajaran mesin yang canggih. Ini melibatkan proses pelatihan model AI menggunakan kumpulan data yang luas dan beragam, termasuk gambar yang dihasilkan oleh berbagai algoritma generatif. Dengan pengolahan data yang mendalam, model ini dapat memahami ciri-ciri unik dari gambar yang dihasilkan oleh AI dan membedakannya dari gambar yang dihasilkan oleh manusia atau alat lain.

Manfaat Utama SynthID:

Akurasi yang Meningkat: Dengan kemampuan mengidentifikasi gambar AI yang semakin canggih, SynthID meningkatkan tingkat akurasi dan keandalan dalam pengenalan gambar. Perlindungan Integritas Konten: Dalam era di mana informasi semakin penting, SynthID membantu dalam melindungi integritas gambar dan konten visual yang dibagikan di platform daring. Katalis Pengembangan AI: Dengan memberikan umpan balik berharga kepada pengembang, SynthID dapat berkontribusi pada pengembangan lebih lanjut dalam bidang kecerdasan buatan. Kesimpulan: Menuju Masa Depan Pengenalan Gambar

Google’s SynthID mewakili langkah penting dalam evolusi kecerdasan buatan. Kemampuan untuk mengenali gambar yang dihasilkan oleh AI memiliki implikasi besar di berbagai industri, mulai dari media hingga keamanan siber. SynthID membuka jalan bagi identifikasi gambar yang lebih akurat dan efektif, membentuk dasar yang lebih kuat untuk penggunaan teknologi AI di masa depan. Dengan langkah ini, Google sekali lagi membuktikan komitmen mereka untuk mendorong batas-batas inovasi dalam dunia teknologi.

© Google Mengenalkan SynthID: Revolusi Identifikasi Gambar Berbasis AI Source: https://www.mpsite.my.id/2023/09/google-mengenalkan-synthid-revolusi.html

[nerdnag]

Today, Google DeepMind and Google Cloud launched SynthID, which is a tool for watermarking and identifying AI-generated images.

The watermark is indiscernible to the naked human eye but can apparently withstand common image manipulation like crops and color adjustments. It isn't foolproof, as stated by Google themselves, but it's "a promising technical approach."

"While generative AI can unlock huge creative potential, it also presents new risks, like enabling creators to spread false information — both intentionally or unintentionally. Being able to identify AI-generated content is critical to empowering people with knowledge of when they’re interacting with generated media, and for helping prevent the spread of misinformation."

My initial thought when I saw this was that in order for this to work, not only do people need to actively choose to watermark their AI images (with Google's tool), it also seems like others who want to find out whether an image has been watermarked need to use the specific, no doubt proprietary technology developed by Google. If this is going to be a useful tool in the long run, there needs to be a standard for what technology generative tools use, and that standard tech then needs to be forcefully applied to all AI-generated imagery regardless of which tool is used - otherwise it's neither practical nor trustworthy. (How are we supposed to trust that an image isn't AI-generated if watermarking it as such is voluntary?)

It's the age-old problem: In order for this to have any effect, it needs to be widely implemented; but it can't be widely implemented if it doesn't exist in the first place, and I suppose we have to start somewhere.

[frank-olivier]

The AI Dilemma: Balancing Benefits and Risks

One of the main focuses of AI research is the development of Artificial General Intelligence (AGI), a hypothetical AI system that surpasses human intelligence in all areas. The AGI timeline, which outlines the expected time frame for the realization of AGI, is a crucial aspect of this research. While some experts predict that AGI will be achieved within the next few years or decades, others argue that it could take centuries or even millennia. Regardless of the time frame, the potential impact of AGI on human society and civilization is enormous and far-reaching.

Another important aspect of AI development is task specialization, where AI models are designed to excel at specific tasks, improving efficiency, productivity, and decision-making. Watermarking technology, which identifies the source of AI-generated content, is also an important part of AI development and addresses concerns about intellectual property and authorship. Google's SynthID technology, which detects and removes AI-generated content on the internet, is another significant development in this field.

However, AI development also brings challenges and concerns. Safety concerns, such as the potential for AI systems to cause harm or injury, must be addressed through robust safety protocols and risk management strategies. Testimonials from whistleblowers and insider perspectives can provide valuable insight into the challenges and successes of AI development and underscore the need for transparency and accountability. Board oversight and governance are also critical to ensure that AI development meets ethical and regulatory standards.

The impact of AI on different industries and aspects of society is also an important consideration. The potential of AI to transform industries such as healthcare, finance and education is enormous, but it also raises concerns about job losses, bias and inequality. The development of AI must be accompanied by a critical examination of its social and economic impacts to ensure that the benefits of AI are distributed fairly and the negative consequences are mitigated.

By recognizing the challenges and complexities of AI development, we can work toward creating a future where AI is developed and deployed in responsible, ethical and beneficial ways.

Ex-OpenAI Employee Reveals Terrifying Future of AI (Matthew Berman, June 2024)

Ex-OpenAI Employees Just Exposed The Truth About AGI (TheAIGRID, October 2024)

Anthropic CEO: AGI is Closer Than You Think [machines of loving grace] (TheAIGRID, October 2024)

AGI in 5 years? Ben Goertzel on Superintelligence (Machine Learning Street Talk, October 2024)

Generative AI and Geopolitical Disruption (Solaris Project, October 2024)

Monday, October 28, 2024

[shivamthakrejr]

Users will have to opt-in to both Labs and the SGE feature to use it, however.

Back in the olden days of last December, we had to go to specialized websites to have our natural language prompts transformed into generated AI art, but no longer! Google announced Thursday that users who have opted-in for its Search Generative Experience (SGE) will be able to create AI images directly from the standard Search bar.

SGE is Google’s vision for our web searching future. Rather than picking websites from a returned list, the system will synthesize a (reasonably) coherent response to the user’s natural language prompt using the same data that the list’s links led to. Thursday’s updates are a natural expansion of that experience, simply returning generated images (using the company’s Imagen text-to-picture AI) instead of generated text. Users type in a description of what they’re looking for (a Capybara cooking breakfast, in Google’s example) and, within moments, the engine will create four alternatives to pick from and refine further. Users will also be able to export their generated images to Drive or download them.

What’s more, users will be able to generate images directly in Google Images. So, if you’re looking for (again, Google’s example) “minimalist halloween table settings” or “spooky dog house ideas,” you’ll be able to type that into the search bar and have Google generate an image based on it. What’s really cool is that you can then turn Google Lens on that generated image to search for actual, real-world products that most closely resemble what the computer hallucinated for you.

There are, of course, a number of limitations built into the new features. You’ll have to be signed up for Google Labs and have opted-in to the SGE program to use any of this. Additionally, the new image generation functions will be available only within the US, in English-language applications and for users over the age of 18. That last requirement is a just bit odd given that Google did just go out of its way to make the program accessible to teens.

The company is also expanding its efforts to rein in the misuse of generative AI technology. Users will be blocked from creating photorealistic images of human faces. You want a photorealistic capybara cooking bacon, that’s no problem. You want a photorealistic Colonel Sanders cooking bacon, you’re going to run into issues and not just in terms of advertising canon. You’ll also be prevented from generating images of “notable” people, so I guess Colonel Sanders is out either way.

Finally, Google is implementing the SynthID system developed by DeepMind announced last month. SythID is a visually undetectable metadata watermark that denotes a generated image as such, as well as provides information on who, or what, created it and when. The new features will be available through the Labs tab (click the flask icon) in the Google app on iOS and Android, and on Chrome desktop to select users today and expanding to more in the coming weeks.

[stablediffusion]

I've had a lot of fun making fakemon (or Synthids as I call them)

Give us a follow on Twitter: @StableDiffusion

H/t m3thlol

[influencermagazineuk]

Google has announced the launch of Veo 2, an enhanced version of its video generation model, alongside updates to Imagen 3 and a new experiment called Whisk, showcasing capabilities with Gemini. First introduced in May at I/O 2024, Veo 2 builds on its predecessor with advancements in understanding “real-world physics and the nuances of human movement and expression”, resulting in greater realism and detail. The new model allows users to specify genre, lens type, and cinematic effects within their prompts. For instance: - “…low-angle tracking shot that glides through the middle of a scene” - “…close-up shot on the face of a scientist looking through her microscope” - “…blur out the background and focus on your subject by putting ‘shallow depth of field’ in your prompt.” If a prompt mentions an “18mm lens”, Veo 2 can produce the distinctive wide-angle shots associated with it. The model also hallucinates “less frequently” and includes the invisible SynthID watermark for added traceability. Veo 2 is being rolled out via VideoFX (part of Google Labs), with Google expanding access to more users, though it remains on a waitlist for now. The company confirmed that Veo 2 will arrive in “YouTube Shorts and other products next year.” “We have been intentionally measured in growing Veo’s availability, so we can help identify, understand and improve the model’s quality and safety while slowly rolling it out via VideoFX, YouTube and Vertex AI,” Google said. Additionally, Google unveiled an updated Imagen 3 model that offers images with “brighter, better composition, richer details and textures” while improving the ability to “render more diverse art styles with greater accuracy.” Imagen 3 is rolling out globally to ImageFX. Lastly, Google introduced “Whisk,” an experimental feature in Google Labs designed to showcase Imagen 3 and Gemini’s visual understanding capabilities. Whisk enables users to prompt using images—one for the subject, another for the scene, and a third for style. These inputs can be remixed to generate unique creations, from digital plushies to enamel pins and stickers. Read the full article

[jcmarchi]

DeepMind unveils Veo 2 model: a new era of video generation?

New Post has been published on https://thedigitalinsider.com/deepmind-unveils-veo-2-model-a-new-era-of-video-generation/

DeepMind unveils Veo 2 model: a new era of video generation?

Just one week after OpenAI released Sora, Google DeepMind has released Veo 2, a vid-gen model pushing hard at the current boundaries of AI-powered video creation.

The model is novel in a number of ways – it’s designed to generate high-quality, 1080p resolution videos that can exceed a minute in length, capturing a wide range of cinematic and visual styles.

Key features & example:

Creates realistic video in phenomenal resolution [up to 4k]

Understands a variety of camera shots [drone, wide, close-up etc]

Better recreates real-world physics & human expression

Prompt: A low-angle shot captures a flock of pink flamingos gracefully wading in a lush, tranquil lagoon. The vibrant pink of their plumage contrasts beautifully with the verdant green of the surrounding vegetation and the crystal-clear turquoise water. Sunlight glints off the water’s surface, creating shimmering reflections that dance on the flamingos’ feathers. The birds’ elegant, curved necks are submerged as they walk through the shallow water, their movements creating gentle ripples that spread across the lagoon. The composition emphasizes the serenity and natural beauty of the scene, highlighting the delicate balance of the ecosystem and the inherent grace of these magnificent birds. The soft, diffused light of early morning bathes the entire scene in a warm, ethereal glow.

[You can explore more prompt & video-generation examples on the official DeepMind release here].

Veo 2 vs Sora; DeepMind vs OpenAI

Veo 2 and OpenAI’s Sora are both groundbreaking AI video generation models, each with its own strengths.

While Sora excels in creative storytelling and imaginative scenarios, Veo 2 prioritizes realism and adherence to real-world physics. Veo 2 also offers a higher degree of control over the video generation process, allowing users to specify camera angles, lighting, and other cinematic elements.

Google’s direct comparison tests, utilizing 1,003 prompts from Meta’s MovieGenBench dataset and human evaluation of 720p, eight-second video clips, revealed Veo 2’s superiority over competitors like OpenAI’s Sora Turbo.

Limitations

While Veo 2 has made significant strides, Google acknowledges the ongoing challenges in consistently generating realistic and dynamic videos, especially in complex scenes and motion sequences.

To mitigate potential misuse and ensure transparency, Veo 2’s initial rollout will be limited to select products like VideoFX, YouTube, and Vertex AI. In 2025, the model’s reach will expand to platforms like YouTube Shorts. All AI-generated videos will be marked with an invisible SynthID watermark.

Other releases

DeepMind also unveiled an enhanced Imagen 3 model, delivering brighter, better-composed images with richer details and textures. This model also excels in rendering diverse art styles with greater accuracy. It is currently being rolled out globally to ImageFX.

Additionally, Google Labs has introduced a new “Whisk” experiment that leverages the updated Imagen 3 and Gemini’s visual understanding capabilities. This experiment allows users to prompt with images, showcasing the advancements in AI-powered image generation.

Like what you see? Then check out tonnes more.

From exclusive content by industry experts and an ever-increasing bank of real world use cases, to 80+ deep-dive summit presentations, our membership plans are packed with awesome AI resources.

Subscribe now

[dapiarim]

Google DeepMind против дезинформации: мнение IT-эксперта, Станислава Дмитриевича Кондрашова

Эта технология, встроенная в ИИ Gemini, предлагает новый подход к борьбе с фейковыми новостями. Вместо обнаружения поддельного контента после публикации (что часто бывает неэффективно), SynthID-Text добавляет незаметные метки прямо в контент во время его создания. Эти метки устойчивы к обработке, такой как перефразирование, перевод или редактирование, и помогают точно определить контент, созданный искусственным интеллектом.

Google сделал исходный код SynthID-Text доступным для других разработчиков и компаний, что является беспрецедентным шагом, учитывая обвинения в неэтичном использовании данных и нарушении прав интеллектуальной собственности.

Система SynthID-Text использует метод модификации вероятностного распределения слов в тексте, сохраняя качество и читаемость текста без ущерба для эффективности маркировки. Google также разрабатывает аналогичные системы для изображений и видео.

Преимущества SynthID-Text включают проактивную защиту, устойчивость к модификациям, открытый исходный код и сохранение качества контента. Однако есть вызовы, такие как возможность обхода системы злоумышленниками и необходимость постоянного улучшения алгоритма.

Станислав Дмитриевич Кондрашов — известный IT-эксперт, который активно следит за развитием искусственного интеллекта (ИИ). Он обеспокоен распространением дезинформации и подделок, которые создаются с помощью современных технологий.

Станислав Дмитриевич выступает за разработку и внедрение новых методов защиты от подобных угроз. Он считает, что необходимо создать систему, которая позволит отслеживать и маркировать контент, созданный с помощью ИИ. Это поможет пользователям отличать настоящие материалы от поддельных.

[aimarketingpoland]

Czy narzędzie Google do znakowania treści pomoże ocalić internet przed zalewem AI?

Google DeepMind, we współpracy z Hugging Face, wprowadziło narzędzie o nazwie SynthID, które ma potencjał do znaczącego wpływu na kontrolę jakości treści generowanych przez sztuczną inteligencję (AI) w internecie. Projekt opiera się na badaniu zatytułowanym „Scalable watermarking for identifying large language model outputs”, które stara się rozróżniać treści tworzone przez ludzi od tych…

[education30and40blog]

Google's AI detection tool can now be tried by anyone

"Google's DeepMind division announced Wednesday that is releasing its SynthID AI content authentication system for general public use ..."

[y2fear]

Google expands Responsible Generative AI Toolkit with support for SynthID, a new Model Alignment library, and more