{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"<iframe src=\"https://embed.acast.com/$/6953b9ead0c0aeaf12bcbd70/6adbbd11b9f7bae151ac1bf9?\" frameBorder=\"0\" width=\"700\" height=\"250\"></iframe>","title":"The Invisible Label: How AI Watermarking Will Change Digital Trust","thumbnail_width":200,"thumbnail_height":200,"thumbnail_url":"https://open-images.acast.com/shows/6953b9ead0c0aeaf12bcbd70/1792785698415-6b9a1060-b36a-4901-98d2-a1f789480d56.jpeg?height=200","description":"<p>AI-generated images, voices, videos and text are becoming harder to distinguish from human-made content. AI watermarking promises to help by embedding an invisible signal that detection software can identify. But what does that signal really prove, and can it survive editing, compression or a determined attacker?</p><p><br></p><p>In this episode of <em>A Beginner’s Guide to AI</em>, Prof. GePhardT examines how AI watermarking works across images, audio, video and text. We separate genuine watermark detection from unreliable attempts to guess whether something “looks like AI.” We also examine why a missing watermark cannot prove that a human created the content.</p><p><br></p><p>The EU AI Act has made this question urgent. Providers of relevant generative AI systems must make synthetic outputs machine-readable and detectable as artificially generated or manipulated. Watermarks are one possible solution, alongside metadata, digital fingerprints and cryptographic content provenance.</p><p><br></p><p>Google SynthID provides our real-world case. The system places imperceptible signals inside supported AI-generated media and allows users to check for them. It demonstrates that watermarking can work at enormous scale, but also reveals the limits of compatibility, robustness and public understanding.</p><p><br></p><h3>🔍 What You Will Learn</h3><ul><li>How invisible AI watermarks are embedded into pixels, audio, video frames and text</li><li>Why watermarking is different from metadata and general AI detection</li><li>What Article 50 of the EU AI Act means for synthetic content</li><li>How Google SynthID and the SynthID Detector work</li><li>Why C2PA and Content Credentials matter</li><li>Whether AI watermarks can be removed</li><li>Why origin and truth are not the same thing</li></ul><p><br></p><p>📧💌📧</p><p>Tune in to get my thoughts and all episodes. Don’t forget to <strong><a href=\"https://beginnersguideto.ai\" rel=\"noopener noreferrer\" target=\"_blank\">subscribe to our newsletter</a></strong>.</p><p>📧💌📧</p><p><br></p><h3>💬 Quotes from the Episode</h3><blockquote>“A detected watermark does not prove that content is false, and the absence of one does not prove that it is human-made.”</blockquote><blockquote>“AI watermarking should be treated as one layer of evidence, not a universal truth machine.”</blockquote><blockquote>“Technology can provide the signature; judgement must still provide the truth.”</blockquote><p><br></p><h3>⏱️ Chapters</h3><p>00:00 Why AI Content Needs an Invisible Label</p><p>02:10 How AI Watermarking Works</p><p>07:25 Watermarks, Detectors and Content Provenance</p><p>10:25 What the EU AI Act Requires</p><p>12:20 Google SynthID in the Real World</p><p>17:15 What Watermarks Can and Cannot Prove </p><p><br></p><h3>👤 About Dietmar Fischer</h3><p>Dietmar is a podcaster and digital marketer from Argo.berlin. If you want to know how to get your AI or digital marketing going, contact him at <strong><a href=\"https://argoberlin.com/\" rel=\"noopener noreferrer\" target=\"_blank\">argoberlin.com</a></strong></p>","author_name":"Dietmar Fischer"}