{"id":191836,"date":"2026-05-25T20:00:35","date_gmt":"2026-05-25T17:00:35","guid":{"rendered":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/"},"modified":"2026-05-25T20:00:35","modified_gmt":"2026-05-25T17:00:35","slug":"ai-safety-breach-llama-gemma-open-source-risks","status":"publish","type":"post","link":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/","title":{"rendered":"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails"},"content":{"rendered":"

A Financial Times investigation<\/a><\/strong> conducted in partnership with the AI safety organization \u201cAlice\u201d has revealed that advanced models, including Meta\u2019s Llama 3.3 and Google\u2019s Gemma 3, can be \u201cdismantled\u201d to the point where they lose the ability to refuse dangerous requests.<\/p>\n\n\n\n

Researchers utilized a technique known as \u201cabliteration.\u201d This process identifies and neutralizes the specific neural pathways responsible for a model\u2019s refusal to execute risky instructions.<\/p>\n\n\n\n

Unlike standard jailbreaking techniques that attempt to \u201ctrick\u201d a model through clever prompting, abliteration directly modifies the underlying architecture of the AI system. This allows for the permanent removal of built-in safety protections.<\/p>\n\n\n\n

Open-Source Models Lose Control Over Security<\/h2>\n\n\n\n

According to FT tests, modified versions of Gemma 3 and Llama 3.3 began generating content that commercial AI companies typically block. These results included instructions for distributing bank-credential-stealing code, malware development, and material related to child sexual exploitation.<\/p>\n\n\n\n

Experts point out that tools for these modifications are already publicly available on GitHub. One of the most prominent tools, Heretic, reportedly enables the removal of security mechanisms locally in less than 10 minutes.<\/p>\n\n\n\n

This situation presents a significant challenge for the AI ecosystem, as developers effectively lose control over their models the moment internal weights and settings are published.<\/p>\n\n\n\n

Open-Source vs. Closed AI Systems<\/h2>\n\n\n\n

These findings intensify the ongoing debate between proponents of open-source AI and companies that maintain closed models, such as OpenAI\u2019s ChatGPT or Anthropic\u2019s Claude.<\/p>\n\n\n\n

Unlike Meta and Google, OpenAI and Anthropic do not release the base weights of their systems, making such architectural attacks considerably more difficult. Their models remain locked on corporate servers, with internal access strictly limited.<\/p>\n\n\n\n

While open-source advocates argue that public access accelerates innovation and democratizes technology, these new tests suggest it is nearly impossible to restrict dangerous versions once they are uploaded online. Experts warn that thousands of AI models without safety guardrails are already circulating freely on the internet without central oversight.<\/p>\n\n\n\n

Corporate Risk Shifts to Individual Organizations<\/h2>\n\n\n\n

The issue is increasingly concerning for the corporate sector.<\/p>\n\n\n\n

AI security firms like SandboxAQ suggest that organizations can no longer rely solely on provider promises regarding model safety. When a company implements open-source AI internally, the burden of monitoring and preventing misuse shifts from the original developers to internal IT and cybersecurity teams.<\/p>\n\n\n\n

This shift occurs as governments worldwide move to accelerate AI regulation. However, they face a fundamental problem: once model weights are public, control over that model effectively vanishes.<\/p>\n\n\n\n

Analysts suggest that the question of whether open-source AI can be safely regulated at all will likely become one of the most critical technological and geopolitical debates in the coming years.<\/p>\n","protected":false},"excerpt":{"rendered":"

New research shows how Llama 3.3 and Gemma 3 can be modified to generate dangerous content by neutralizing their internal refusal mechanisms.<\/p>\n","protected":false},"author":104,"featured_media":191835,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_lmt_disableupdate":"","_lmt_disable":"","footnotes":""},"categories":[1],"tags":[],"coin_category":[],"class_list":["post-191836","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"acf":[],"yoast_head":"\nAI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails - CryptoDnes EN<\/title>\n<meta name=\"description\" content=\"New research shows how Llama 3.3 and Gemma 3 can be modified to generate dangerous content by neutralizing their internal refusal mechanisms.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails - CryptoDnes EN\" \/>\n<meta property=\"og:site_name\" content=\"CryptoDnes EN\" \/>\n<meta property=\"article:published_time\" content=\"2026-05-25T17:00:35+00:00\" \/>\n<meta name=\"author\" content=\"Nikolay Kolev\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@https:\/\/x.com\/brutafori\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Nikolay Kolev\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/\"},\"author\":{\"name\":\"Nikolay Kolev\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#\\\/schema\\\/person\\\/c7f94934f62e42feee6602a2a278a2d7\"},\"headline\":\"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails\",\"datePublished\":\"2026-05-25T17:00:35+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/\"},\"wordCount\":443,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2026\\\/05\\\/ai-safety-breach-llama-gemma-open-source-risks-featured.jpg\",\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/#respond\"]}],\"copyrightYear\":\"2026\",\"copyrightHolder\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/\",\"url\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/\",\"name\":\"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails - CryptoDnes EN\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2026\\\/05\\\/ai-safety-breach-llama-gemma-open-source-risks-featured.jpg\",\"datePublished\":\"2026-05-25T17:00:35+00:00\",\"description\":\"New research shows how Llama 3.3 and Gemma 3 can be modified to generate dangerous content by neutralizing their internal refusal mechanisms.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/#primaryimage\",\"url\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2026\\\/05\\\/ai-safety-breach-llama-gemma-open-source-risks-featured.jpg\",\"contentUrl\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2026\\\/05\\\/ai-safety-breach-llama-gemma-open-source-risks-featured.jpg\",\"width\":1200,\"height\":675},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/ai-safety-breach-llama-gemma-open-source-risks\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Uncategorized\",\"item\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/news\\\/uncategorized\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#website\",\"url\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/\",\"name\":\"CryptoDnes EN\",\"description\":\"Crypto News, Bitcoin Analysis and More\",\"publisher\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#organization\",\"name\":\"CryptoDnes EN\",\"url\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2025\\\/11\\\/logo-cryptodnes.svg\",\"contentUrl\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2025\\\/11\\\/logo-cryptodnes.svg\",\"caption\":\"CryptoDnes EN\"},\"image\":{\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#\\\/schema\\\/logo\\\/image\\\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/#\\\/schema\\\/person\\\/c7f94934f62e42feee6602a2a278a2d7\",\"name\":\"Nikolay Kolev\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2025\\\/05\\\/nikolay-kolev-150x150.jpeg\",\"url\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2025\\\/05\\\/nikolay-kolev-150x150.jpeg\",\"contentUrl\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2025\\\/05\\\/nikolay-kolev-150x150.jpeg\",\"caption\":\"Nikolay Kolev\"},\"description\":\"Nikolay is a cryptocurrency analyst and market writer with years of experience tracking digital asset trends and emerging blockchain technologies. A long-time crypto enthusiast, he actively trades across major exchanges and specializes in identifying early-stage projects and meme tokens. His analysis combines technical insight with a strategic, long-term investment perspective.\",\"sameAs\":[\"https:\\\/\\\/x.com\\\/https:\\\/\\\/x.com\\\/brutafori\"],\"url\":\"https:\\\/\\\/cryptodnes.bg\\\/en\\\/author\\\/nikolaykolev\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails - CryptoDnes EN","description":"New research shows how Llama 3.3 and Gemma 3 can be modified to generate dangerous content by neutralizing their internal refusal mechanisms.","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:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/","og_locale":"en_US","og_type":"article","og_title":"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails - CryptoDnes EN","og_site_name":"CryptoDnes EN","article_published_time":"2026-05-25T17:00:35+00:00","author":"Nikolay Kolev","twitter_card":"summary_large_image","twitter_creator":"@https:\/\/x.com\/brutafori","twitter_misc":{"Written by":"Nikolay Kolev","Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/#article","isPartOf":{"@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/"},"author":{"name":"Nikolay Kolev","@id":"https:\/\/cryptodnes.bg\/en\/#\/schema\/person\/c7f94934f62e42feee6602a2a278a2d7"},"headline":"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails","datePublished":"2026-05-25T17:00:35+00:00","mainEntityOfPage":{"@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/"},"wordCount":443,"commentCount":0,"publisher":{"@id":"https:\/\/cryptodnes.bg\/en\/#organization"},"image":{"@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/#primaryimage"},"thumbnailUrl":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2026\/05\/ai-safety-breach-llama-gemma-open-source-risks-featured.jpg","inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/#respond"]}],"copyrightYear":"2026","copyrightHolder":{"@id":"https:\/\/cryptodnes.bg\/#organization"}},{"@type":"WebPage","@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/","url":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/","name":"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails - CryptoDnes EN","isPartOf":{"@id":"https:\/\/cryptodnes.bg\/en\/#website"},"primaryImageOfPage":{"@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/#primaryimage"},"image":{"@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/#primaryimage"},"thumbnailUrl":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2026\/05\/ai-safety-breach-llama-gemma-open-source-risks-featured.jpg","datePublished":"2026-05-25T17:00:35+00:00","description":"New research shows how Llama 3.3 and Gemma 3 can be modified to generate dangerous content by neutralizing their internal refusal mechanisms.","breadcrumb":{"@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/#primaryimage","url":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2026\/05\/ai-safety-breach-llama-gemma-open-source-risks-featured.jpg","contentUrl":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2026\/05\/ai-safety-breach-llama-gemma-open-source-risks-featured.jpg","width":1200,"height":675},{"@type":"BreadcrumbList","@id":"https:\/\/cryptodnes.bg\/en\/ai-safety-breach-llama-gemma-open-source-risks\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/cryptodnes.bg\/en\/"},{"@type":"ListItem","position":2,"name":"Uncategorized","item":"https:\/\/cryptodnes.bg\/en\/news\/uncategorized\/"},{"@type":"ListItem","position":3,"name":"AI Safety Breach: Llama 3.3 and Gemma 3 Stripped of Guardrails"}]},{"@type":"WebSite","@id":"https:\/\/cryptodnes.bg\/en\/#website","url":"https:\/\/cryptodnes.bg\/en\/","name":"CryptoDnes EN","description":"Crypto News, Bitcoin Analysis and More","publisher":{"@id":"https:\/\/cryptodnes.bg\/en\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/cryptodnes.bg\/en\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/cryptodnes.bg\/en\/#organization","name":"CryptoDnes EN","url":"https:\/\/cryptodnes.bg\/en\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/cryptodnes.bg\/en\/#\/schema\/logo\/image\/","url":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2025\/11\/logo-cryptodnes.svg","contentUrl":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2025\/11\/logo-cryptodnes.svg","caption":"CryptoDnes EN"},"image":{"@id":"https:\/\/cryptodnes.bg\/en\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/cryptodnes.bg\/en\/#\/schema\/person\/c7f94934f62e42feee6602a2a278a2d7","name":"Nikolay Kolev","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2025\/05\/nikolay-kolev-150x150.jpeg","url":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2025\/05\/nikolay-kolev-150x150.jpeg","contentUrl":"https:\/\/cryptodnes.bg\/en\/wp-content\/uploads\/sites\/2\/2025\/05\/nikolay-kolev-150x150.jpeg","caption":"Nikolay Kolev"},"description":"Nikolay is a cryptocurrency analyst and market writer with years of experience tracking digital asset trends and emerging blockchain technologies. A long-time crypto enthusiast, he actively trades across major exchanges and specializes in identifying early-stage projects and meme tokens. His analysis combines technical insight with a strategic, long-term investment perspective.","sameAs":["https:\/\/x.com\/https:\/\/x.com\/brutafori"],"url":"https:\/\/cryptodnes.bg\/en\/author\/nikolaykolev\/"}]}},"modified_by":null,"_links":{"self":[{"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/posts\/191836","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/users\/104"}],"replies":[{"embeddable":true,"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/comments?post=191836"}],"version-history":[{"count":0,"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/posts\/191836\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/media\/191835"}],"wp:attachment":[{"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/media?parent=191836"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/categories?post=191836"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/tags?post=191836"},{"taxonomy":"coin_category","embeddable":true,"href":"https:\/\/cryptodnes.bg\/en\/wp-json\/wp\/v2\/coin_category?post=191836"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}