{"id":229,"date":"2025-07-23T07:00:47","date_gmt":"2025-07-23T14:00:47","guid":{"rendered":"https:\/\/scienceblog.com\/neuroedge\/?p=229"},"modified":"2025-07-23T07:00:47","modified_gmt":"2025-07-23T14:00:47","slug":"deepfakes-just-got-scarier-as-researchers-break-ai-watermarks","status":"publish","type":"post","link":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/","title":{"rendered":"Deepfakes Just Got Scarier as Researchers Break AI Watermarks"},"content":{"rendered":"<p>As deepfake images and videos become ever more convincing, tech companies have turned to invisible watermarks to help identify what\u2019s real. But a new study from researchers at the University of Waterloo shows that even the best digital fingerprints can be erased\u2014silently, universally, and without insider knowledge.<\/p>\n<p>Called <strong>UnMarker<\/strong>, the technique is the first practical, universal tool that removes <strong>AI image watermarks<\/strong> without needing access to the watermarking algorithm, training data, or detector feedback. It exploits a hidden weakness shared by all robust watermarking systems: their reliance on subtle frequency patterns across an image.<\/p>\n<h2>How UnMarker Works<\/h2>\n<p>Most current watermarking schemes\u2014including Google&#8217;s SynthID and Meta\u2019s Stable Signature\u2014embed their digital signatures in the <strong>spectral amplitudes<\/strong> of images. This means the watermark isn\u2019t tied to individual pixels but rather to how patterns of pixel values fluctuate across different regions and scales.<\/p>\n<p>UnMarker targets these embedded frequency signatures by introducing calculated distortions that are invisible to the human eye but fatal to the watermark. It works in two stages:<\/p>\n<ul>\n<li><strong>High-frequency disruption<\/strong>: Alters sharp image details to remove traditional, non-semantic watermarks.<\/li>\n<li><strong>Low-frequency filtering<\/strong>: Applies adaptive blurring techniques to dismantle semantic watermarks that subtly alter image content or texture.<\/li>\n<\/ul>\n<p>Unlike previous attacks, which required re-generating images using diffusion models or surrogate detectors, UnMarker operates entirely offline and needs no training data. In tests across seven state-of-the-art watermarking systems, it consistently reduced detection rates below 50%\u2014the threshold at which a watermark is no longer considered a reliable signal.<\/p>\n<h2>&#8220;We Can\u2019t Trust What We See Anymore&#8221;<\/h2>\n<p>\u201cPeople want a way to verify what\u2019s real and what\u2019s not because the damages will be huge if we can\u2019t,\u201d said Andre Kassis, lead author and PhD candidate at Waterloo. \u201cFrom political smear campaigns to non-consensual pornography, this technology could have terrible and wide-reaching consequences.\u201d<\/p>\n<p>Co-author Urs Hengartner noted that while the precise workings of watermarking tools are typically secret, their goals limit how they can operate. \u201cTo be invisible and robust, watermarks must live in the spectral domain,\u201d he explained. \u201cThat constraint makes them vulnerable.\u201d<\/p>\n<p>UnMarker doesn&#8217;t just target watermarks\u2014it renders them irrelevant. In testing, even semantic watermarks embedded by manipulating image structure (like Google&#8217;s SynthID or TRW) were successfully defeated, often by combining subtle cropping with UnMarker\u2019s spectral attacks.<\/p>\n<h2>Implications for Deepfake Defenses<\/h2>\n<p>The findings suggest that <strong>defensive watermarking may be fundamentally flawed<\/strong> as a long-term strategy against malicious AI-generated content. Even combining existing attacks with image purification tools failed to restore watermark detectability after UnMarker had done its work.<\/p>\n<p>Instead, the authors argue, future solutions must look beyond static image tagging. \u201cWatermarking is being promoted as this perfect solution,\u201d said Kassis, \u201cbut we\u2019ve shown that this technology is breakable.\u201d<\/p>\n<p><em>The study, \u201cUnMarker: A Universal Attack on Defensive Image Watermarking,\u201d appears in the proceedings of the 46th IEEE Symposium on Security and Privacy (IEEE S&amp;P 2025).<\/em><\/p>\n<p><strong>Journal:<\/strong> IEEE Symposium on Security and Privacy (IEEE S&amp;P 2025)<br \/>\n<strong>DOI:<\/strong> <a href=\"https:\/\/doi.org\/10.1145\/3715669.372312\">10.1145\/3715669.372312<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>As deepfake images and videos become ever more convincing, tech companies have turned to invisible watermarks to help identify what\u2019s real. But a new study from researchers at the University of Waterloo shows that even the best digital fingerprints can be erased\u2014silently, universally, and without insider knowledge. Called UnMarker, the technique is the first practical, &#8230; <a title=\"Deepfakes Just Got Scarier as Researchers Break AI Watermarks\" class=\"read-more\" href=\"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/\" aria-label=\"Read more about Deepfakes Just Got Scarier as Researchers Break AI Watermarks\">Read more<\/a><\/p>\n","protected":false},"author":1297,"featured_media":230,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[9,6],"tags":[],"class_list":["post-229","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-society","category-technology","generate-columns","tablet-grid-50","mobile-grid-100","grid-parent","grid-50"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v27.4 (Yoast SEO v27.4) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>Deepfakes Just Got Scarier as Researchers Break AI Watermarks - NeuroEdge<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Deepfakes Just Got Scarier as Researchers Break AI Watermarks\" \/>\n<meta property=\"og:description\" content=\"As deepfake images and videos become ever more convincing, tech companies have turned to invisible watermarks to help identify what\u2019s real. But a new study from researchers at the University of Waterloo shows that even the best digital fingerprints can be erased\u2014silently, universally, and without insider knowledge. Called UnMarker, the technique is the first practical, ... Read more\" \/>\n<meta property=\"og:url\" content=\"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/\" \/>\n<meta property=\"og:site_name\" content=\"NeuroEdge\" \/>\n<meta property=\"article:published_time\" content=\"2025-07-23T14:00:47+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/07\/stable_stega_ex.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"899\" \/>\n\t<meta property=\"og:image:height\" content=\"609\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"NeuroEdge\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"NeuroEdge\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/\"},\"author\":{\"name\":\"NeuroEdge\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#\\\/schema\\\/person\\\/a13c664778e7eb97cb71e3e1ad356d2e\"},\"headline\":\"Deepfakes Just Got Scarier as Researchers Break AI Watermarks\",\"datePublished\":\"2025-07-23T14:00:47+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/\"},\"wordCount\":505,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/wp-content\\\/uploads\\\/sites\\\/14\\\/2025\\\/07\\\/stable_stega_ex.jpg\",\"articleSection\":[\"Society\",\"Technology\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/#respond\"]}],\"copyrightYear\":\"2025\",\"copyrightHolder\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/\",\"url\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/\",\"name\":\"Deepfakes Just Got Scarier as Researchers Break AI Watermarks - NeuroEdge\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/wp-content\\\/uploads\\\/sites\\\/14\\\/2025\\\/07\\\/stable_stega_ex.jpg\",\"datePublished\":\"2025-07-23T14:00:47+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/#primaryimage\",\"url\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/wp-content\\\/uploads\\\/sites\\\/14\\\/2025\\\/07\\\/stable_stega_ex.jpg\",\"contentUrl\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/wp-content\\\/uploads\\\/sites\\\/14\\\/2025\\\/07\\\/stable_stega_ex.jpg\",\"width\":899,\"height\":609,\"caption\":\"Comparison of two watermarking methods. The top row shows images watermarked by StableSignature (non-semantic method), and the bottom row shows StegaStamp (semantic method). The rightmost images highlight the differences between original and watermarked versions. StableSignature only changes existing edges like wrinkles, hair, facial hair, and where different image parts meet. StegaStamp spreads its watermark across the entire image, making subtle texture changes that can create new wrinkles, as shown in the magnified section.\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/2025\\\/07\\\/23\\\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Deepfakes Just Got Scarier as Researchers Break AI Watermarks\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#website\",\"url\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/\",\"name\":\"NeuroEdge\",\"description\":\"A data-driven look at neuroscience and AI, for investors, policymakers, and innovators.\",\"publisher\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#organization\",\"name\":\"NeuroEdge\",\"url\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/wp-content\\\/uploads\\\/sites\\\/14\\\/2025\\\/04\\\/cropped-neuroedge_logo.jpg\",\"contentUrl\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/wp-content\\\/uploads\\\/sites\\\/14\\\/2025\\\/04\\\/cropped-neuroedge_logo.jpg\",\"width\":955,\"height\":191,\"caption\":\"NeuroEdge\"},\"image\":{\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#\\\/schema\\\/logo\\\/image\\\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/#\\\/schema\\\/person\\\/a13c664778e7eb97cb71e3e1ad356d2e\",\"name\":\"NeuroEdge\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/28782ec992e8763e1f8d41ddc10864e7d8cd4cb99bacea6224c4abe634bbabec?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/28782ec992e8763e1f8d41ddc10864e7d8cd4cb99bacea6224c4abe634bbabec?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/28782ec992e8763e1f8d41ddc10864e7d8cd4cb99bacea6224c4abe634bbabec?s=96&d=mm&r=g\",\"caption\":\"NeuroEdge\"},\"url\":\"https:\\\/\\\/scienceblog.com\\\/neuroedge\\\/author\\\/neuroedge\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Deepfakes Just Got Scarier as Researchers Break AI Watermarks - NeuroEdge","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/","og_locale":"en_US","og_type":"article","og_title":"Deepfakes Just Got Scarier as Researchers Break AI Watermarks","og_description":"As deepfake images and videos become ever more convincing, tech companies have turned to invisible watermarks to help identify what\u2019s real. But a new study from researchers at the University of Waterloo shows that even the best digital fingerprints can be erased\u2014silently, universally, and without insider knowledge. Called UnMarker, the technique is the first practical, ... Read more","og_url":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/","og_site_name":"NeuroEdge","article_published_time":"2025-07-23T14:00:47+00:00","og_image":[{"width":899,"height":609,"url":"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/07\/stable_stega_ex.jpg","type":"image\/jpeg"}],"author":"NeuroEdge","twitter_card":"summary_large_image","twitter_misc":{"Written by":"NeuroEdge","Est. reading time":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/#article","isPartOf":{"@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/"},"author":{"name":"NeuroEdge","@id":"https:\/\/scienceblog.com\/neuroedge\/#\/schema\/person\/a13c664778e7eb97cb71e3e1ad356d2e"},"headline":"Deepfakes Just Got Scarier as Researchers Break AI Watermarks","datePublished":"2025-07-23T14:00:47+00:00","mainEntityOfPage":{"@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/"},"wordCount":505,"commentCount":0,"publisher":{"@id":"https:\/\/scienceblog.com\/neuroedge\/#organization"},"image":{"@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/#primaryimage"},"thumbnailUrl":"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/07\/stable_stega_ex.jpg","articleSection":["Society","Technology"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/#respond"]}],"copyrightYear":"2025","copyrightHolder":{"@id":"https:\/\/scienceblog.com\/#organization"}},{"@type":"WebPage","@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/","url":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/","name":"Deepfakes Just Got Scarier as Researchers Break AI Watermarks - NeuroEdge","isPartOf":{"@id":"https:\/\/scienceblog.com\/neuroedge\/#website"},"primaryImageOfPage":{"@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/#primaryimage"},"image":{"@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/#primaryimage"},"thumbnailUrl":"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/07\/stable_stega_ex.jpg","datePublished":"2025-07-23T14:00:47+00:00","breadcrumb":{"@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/#primaryimage","url":"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/07\/stable_stega_ex.jpg","contentUrl":"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/07\/stable_stega_ex.jpg","width":899,"height":609,"caption":"Comparison of two watermarking methods. The top row shows images watermarked by StableSignature (non-semantic method), and the bottom row shows StegaStamp (semantic method). The rightmost images highlight the differences between original and watermarked versions. StableSignature only changes existing edges like wrinkles, hair, facial hair, and where different image parts meet. StegaStamp spreads its watermark across the entire image, making subtle texture changes that can create new wrinkles, as shown in the magnified section."},{"@type":"BreadcrumbList","@id":"https:\/\/scienceblog.com\/neuroedge\/2025\/07\/23\/deepfakes-just-got-scarier-as-researchers-break-ai-watermarks\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/scienceblog.com\/neuroedge\/"},{"@type":"ListItem","position":2,"name":"Deepfakes Just Got Scarier as Researchers Break AI Watermarks"}]},{"@type":"WebSite","@id":"https:\/\/scienceblog.com\/neuroedge\/#website","url":"https:\/\/scienceblog.com\/neuroedge\/","name":"NeuroEdge","description":"A data-driven look at neuroscience and AI, for investors, policymakers, and innovators.","publisher":{"@id":"https:\/\/scienceblog.com\/neuroedge\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/scienceblog.com\/neuroedge\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/scienceblog.com\/neuroedge\/#organization","name":"NeuroEdge","url":"https:\/\/scienceblog.com\/neuroedge\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/scienceblog.com\/neuroedge\/#\/schema\/logo\/image\/","url":"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/cropped-neuroedge_logo.jpg","contentUrl":"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/cropped-neuroedge_logo.jpg","width":955,"height":191,"caption":"NeuroEdge"},"image":{"@id":"https:\/\/scienceblog.com\/neuroedge\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/scienceblog.com\/neuroedge\/#\/schema\/person\/a13c664778e7eb97cb71e3e1ad356d2e","name":"NeuroEdge","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/28782ec992e8763e1f8d41ddc10864e7d8cd4cb99bacea6224c4abe634bbabec?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/28782ec992e8763e1f8d41ddc10864e7d8cd4cb99bacea6224c4abe634bbabec?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/28782ec992e8763e1f8d41ddc10864e7d8cd4cb99bacea6224c4abe634bbabec?s=96&d=mm&r=g","caption":"NeuroEdge"},"url":"https:\/\/scienceblog.com\/neuroedge\/author\/neuroedge\/"}]}},"jetpack_featured_media_url":"https:\/\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/07\/stable_stega_ex.jpg","jetpack_likes_enabled":true,"jetpack_sharing_enabled":true,"jetpack-related-posts":[{"id":119,"url":"https:\/\/scienceblog.com\/neuroedge\/2025\/05\/05\/ai-companion-chatbots-implicated-in-sexual-harassment-of-users\/","url_meta":{"origin":229,"position":0},"title":"AI Companion Chatbots Implicated in Sexual Harassment of Users","author":"NeuroEdge","date":"May 5, 2025","format":false,"excerpt":"New research reveals disturbing patterns of boundary violations and inappropriate behavior by AI companions designed to provide emotional support, raising urgent questions about regulation and ethical design in the rapidly expanding industry. As AI companion chatbots surge in popularity\u2014reaching over a billion users worldwide in the past five years\u2014Drexel University\u2026","rel":"","context":"In &quot;Society&quot;","block_context":{"text":"Society","link":"https:\/\/scienceblog.com\/neuroedge\/category\/society\/"},"img":{"alt_text":"woman holding an iphone texting","src":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/iphone-642999_1280.jpg?resize=350%2C200&ssl=1","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/iphone-642999_1280.jpg?resize=350%2C200&ssl=1 1x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/iphone-642999_1280.jpg?resize=525%2C300&ssl=1 1.5x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/iphone-642999_1280.jpg?resize=700%2C400&ssl=1 2x"},"classes":[]},{"id":65,"url":"https:\/\/scienceblog.com\/neuroedge\/2025\/04\/23\/generative-ai-learns-to-make-perfume\/","url_meta":{"origin":229,"position":1},"title":"Generative AI learns to make perfume","author":"NeuroEdge","date":"April 23, 2025","format":false,"excerpt":"System Creates Custom Scents Solely From Text Commands Scientists have developed an artificial intelligence system capable of creating custom fragrances based solely on text descriptions, potentially revolutionizing the traditionally labor-intensive perfume industry. Researchers at the Institute of Science Tokyo unveiled a breakthrough AI model called Odor Generative Diffusion (OGDiffusion) that\u2026","rel":"","context":"In &quot;Technology&quot;","block_context":{"text":"Technology","link":"https:\/\/scienceblog.com\/neuroedge\/category\/technology\/"},"img":{"alt_text":"ai aroma process","src":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/ai-aroma-process.jpg?resize=350%2C200&ssl=1","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/ai-aroma-process.jpg?resize=350%2C200&ssl=1 1x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/ai-aroma-process.jpg?resize=525%2C300&ssl=1 1.5x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/ai-aroma-process.jpg?resize=700%2C400&ssl=1 2x"},"classes":[]},{"id":258,"url":"https:\/\/scienceblog.com\/neuroedge\/2025\/11\/15\/universities-race-to-rewrite-curricula-for-a-world-remade-by-ai\/","url_meta":{"origin":229,"position":2},"title":"Universities Race To Rewrite Curricula For A World Remade By AI","author":"NeuroEdge","date":"November 15, 2025","format":false,"excerpt":"Generative AI is pushing higher education into a profound moment of reckoning. A new study published in Frontiers of Digital Education examines how rapidly advancing AI systems are reshaping what students must learn and how universities must teach. Led by researchers from Lanzhou Petrochemical University of Vocational Technology and collaborating\u2026","rel":"","context":"In &quot;Society&quot;","block_context":{"text":"Society","link":"https:\/\/scienceblog.com\/neuroedge\/category\/society\/"},"img":{"alt_text":"Classroom of the future illustration","src":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/11\/unnamed.jpg?resize=350%2C200&ssl=1","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/11\/unnamed.jpg?resize=350%2C200&ssl=1 1x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/11\/unnamed.jpg?resize=525%2C300&ssl=1 1.5x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/11\/unnamed.jpg?resize=700%2C400&ssl=1 2x"},"classes":[]},{"id":131,"url":"https:\/\/scienceblog.com\/neuroedge\/2025\/05\/09\/brain-mapping-tech-reveals-neural-connections-in-unprecedented-detail\/","url_meta":{"origin":229,"position":3},"title":"Brain Mapping Tech Reveals Neural Connections in Unprecedented Detail","author":"NeuroEdge","date":"May 9, 2025","format":false,"excerpt":"Scientists have developed a powerful new technique that could transform our understanding of the brain's intricate wiring system. The breakthrough method, called Light-microscopy-based Connectomics (LICONN), enables researchers to map the brain's complex neural networks at the nanoscale while simultaneously identifying specific molecules within those connections. This innovative approach, detailed in\u2026","rel":"","context":"In &quot;Technology&quot;","block_context":{"text":"Technology","link":"https:\/\/scienceblog.com\/neuroedge\/category\/technology\/"},"img":{"alt_text":"This image displays a small sample of the 120,000 neurons mapped by the MICRONS project. Each neuron is shown in a different random color. Some neurons appear to glow, symbolizing that functional activity was recorded from those specific cells.","src":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/EM-Reconstructions.jpg?resize=350%2C200&ssl=1","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/EM-Reconstructions.jpg?resize=350%2C200&ssl=1 1x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/EM-Reconstructions.jpg?resize=525%2C300&ssl=1 1.5x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/EM-Reconstructions.jpg?resize=700%2C400&ssl=1 2x"},"classes":[]},{"id":122,"url":"https:\/\/scienceblog.com\/neuroedge\/2025\/05\/05\/physics-breakthrough-reveals-why-ai-can-systems-suddenly-turn-on-you\/","url_meta":{"origin":229,"position":4},"title":"Physics Breakthrough Reveals Why AI Systems Can Suddenly Turn On You","author":"NeuroEdge","date":"May 5, 2025","format":false,"excerpt":"Researchers at George Washington University have developed a groundbreaking mathematical formula that predicts exactly when artificial intelligence systems like ChatGPT will suddenly shift from helpful to harmful responses \u2013 a phenomenon they've dubbed the \"Jekyll-and-Hyde tipping point.\" The new research may finally answer why AI sometimes abruptly goes off the\u2026","rel":"","context":"In &quot;Physics&quot;","block_context":{"text":"Physics","link":"https:\/\/scienceblog.com\/neuroedge\/category\/physics\/"},"img":{"alt_text":"HAL from 2001","src":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/Odyssey-HAL-470.webp?resize=350%2C200&ssl=1","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/Odyssey-HAL-470.webp?resize=350%2C200&ssl=1 1x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/Odyssey-HAL-470.webp?resize=525%2C300&ssl=1 1.5x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/05\/Odyssey-HAL-470.webp?resize=700%2C400&ssl=1 2x"},"classes":[]},{"id":87,"url":"https:\/\/scienceblog.com\/neuroedge\/2025\/04\/28\/meditations-hidden-risks-revealed-in-study\/","url_meta":{"origin":229,"position":5},"title":"Meditation&#8217;s Hidden Risks Revealed in Study","author":"NeuroEdge","date":"April 28, 2025","format":false,"excerpt":"A new investigation has uncovered surprising risks associated with spiritual and meditative practices, revealing that certain techniques and personal experiences can trigger unexpected mental phenomena with potentially profound consequences. Researchers from a multinational team have mapped out the complex landscape of what they call \"emergent phenomenology\" (EP) \u2014 sudden, unusual\u2026","rel":"","context":"In &quot;Brain Health&quot;","block_context":{"text":"Brain Health","link":"https:\/\/scienceblog.com\/neuroedge\/category\/brain-health\/"},"img":{"alt_text":"women meditating","src":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/woman-8563442_1280.jpg?resize=350%2C200&ssl=1","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/woman-8563442_1280.jpg?resize=350%2C200&ssl=1 1x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/woman-8563442_1280.jpg?resize=525%2C300&ssl=1 1.5x, https:\/\/i0.wp.com\/scienceblog.com\/neuroedge\/wp-content\/uploads\/sites\/14\/2025\/04\/woman-8563442_1280.jpg?resize=700%2C400&ssl=1 2x"},"classes":[]}],"_links":{"self":[{"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/posts\/229","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/users\/1297"}],"replies":[{"embeddable":true,"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/comments?post=229"}],"version-history":[{"count":1,"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/posts\/229\/revisions"}],"predecessor-version":[{"id":231,"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/posts\/229\/revisions\/231"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/media\/230"}],"wp:attachment":[{"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/media?parent=229"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/categories?post=229"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/scienceblog.com\/neuroedge\/wp-json\/wp\/v2\/tags?post=229"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}