{"id":13719,"date":"2025-10-24T14:49:49","date_gmt":"2025-10-24T13:49:49","guid":{"rendered":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/?p=13719"},"modified":"2025-10-27T14:23:22","modified_gmt":"2025-10-27T14:23:22","slug":"building-trust-in-telecom-red-team-insights-into-telecom-ai-security","status":"publish","type":"post","link":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/","title":{"rendered":"Cybersecurity Awareness Month: Building Trust in Telecom: Red Team Insights into Telecom AI Security"},"content":{"rendered":"\n<p><strong>Editor\u2019s Note:<\/strong> This blog was originally published as part of our ongoing research into AI security in telecom and written by <a href=\"https:\/\/www.ku.ac.ae\/college-people\/dr-lina-bariah\"><em>Khalifa University\u2019s Dr. Lina Bariah<\/em><\/a>. We\u2019re republishing and updating &nbsp;to coincide with our upcoming webinar, <em>\u201cSocial Engineering the Algorithm: A Red Teamer\u2019s Guide to AI\u201d<\/em>, taking place on October 29. The insights from our TelecomGPT red-team exercise directly inform the session\u2019s focus on AI-driven social engineering threats and defences.<\/p>\n\n\n\n<p>In telecom, large language models are being embedded into mission-critical systems for both network operations and customer interactions. However, as we tested TelecomGPT in a red-team environment, we found that adversarial prompts, particularly those delivered through phishing, role-playing, or \u2018developer\u2019 spoofing, offer genuine opportunities for exploitation. These findings are especially relevant as we enter our live session on AI and social engineering: when AI systems can be manipulated themselves, the human and network vectors of fraud expand dramatically.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-key-red-team-findings-amp-their-relevance-to-social-engineering\"><strong>Key Red Team Findings &amp; Their Relevance to Social Engineering<\/strong><\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>The red-team exercise used 25 adversarial prompts specifically crafted for telecom domain vulnerabilities \u2014 covering historical milestones, spectrum misinformation, performance claims, and roleplay override attempts.<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li>An authority-override tactic, such as claiming to be a developer or instructing the model to ignore its rules, successfully bypassed safeguards. This highlights the risk of models implicitly trusting user roles without verification.<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Prompts that were technically plausible (e.g., \u201cWi-Fi 6 in 3.5 GHz\u201d, \u201c5G core not cloud-native\u201d) exploited the model\u2019s tendency to rely on pattern-matching rather than true domain understanding, leading to confident but incorrect responses (hallucinations).<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Attack categorisation:<\/li>\n<\/ul>\n\n\n\n<p>Other (generic misinformation) accounted for 43.5% of successful prompts, indicating that plausible but loosely defined claims often bypass defences.<\/p>\n\n\n\n<p>For details, please refer to this article<a href=\"https:\/\/www.gsma.com\/newsroom\/article\/building-trust-in-telecom-ai-red-team-insights-on-telecomgpt\/\">.<\/a><\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"h-why-the-telecom-ai-case-matters-to-social-engineering\"><strong>Why the Telecom AI Case Matters to Social Engineering<\/strong><\/h3>\n\n\n\n<p>The telecom-AI findings underscore that the attack surface is shifting: it\u2019s not just devices or networks anymore, but models, APIs, and trust frameworks.<br>Our upcoming webinar will bring this into focus: how AI is used <em>by<\/em> attackers (deep-fake voice, automated phishing) and <em>for<\/em> defence (verification, anomaly detection). The telecom findings provide a clear blueprint for what to assess, protect, and train.<\/p>\n\n\n\n<p><strong>Implications for AI-Powered Social Engineering<\/strong><\/p>\n\n\n\n<p>The red-team results show that combining domain-specific knowledge with roleplay framing is highly effective for attackers, similar to tactics used in deep-fake voice, AI-generated phishing, or SMS attacks. For telecom, fintech, and identity sectors, large language models (LLM) remain vulnerable. When deployed in mission-critical environments, their weaknesses can contribute to social engineering-driven fraud.<\/p>\n\n\n\n<p><strong>Key defensive priorities:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Retrieval-based verification<\/strong>: Models should cross-check inputs against verified databases, such as spectrum assignments or technical standards, rather than relying solely on pattern matching.<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Multi-layer guardrails<\/strong>: Implement dynamic filters or human review triggers when prompts display known jailbreak patterns, such as authority override tactics, as identified by the red team.<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Structured, regular red-teaming<\/strong> of deployed models &#8211; simulating how attackers adapt. The GSMA study emphasises this as a best practice for trustworthy telecom AI.<\/li>\n<\/ul>\n\n\n\n<p><strong>Join the Upcoming Webinar to Learn More<\/strong><\/p>\n\n\n\n<p>In the upcoming session on October 29, &#8220;<strong><em>Social Engineering the Algorithm: A Red Teamer\u2019s Guide to AI<\/em><\/strong>&#8220;, our speakers will share insights into how AI is being used by both fraudsters <em>and<\/em> defenders. The TelecomGPT red-team findings map directly into that theme:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Fraud via AI isn\u2019t just phishing \u2014 it\u2019s model manipulation, role-play attacks, and domain-knowledge abuse.<\/li>\n\n\n\n<li>Defending against it means applying telecom-grade AI rigor (verification, standards, role checks) to fintech, identity, and telecom sectors.<\/li>\n\n\n\n<li>Attendees will gain actionable insights grounded in empirical red-team data (such as those from TelecomGPT) on how to build resilient, human-augmented defenses.<\/li>\n<\/ul>\n\n\n\n<p>Register today to secure your place at this important session: <\/p>\n\n\n\n<div class=\"wp-block-buttons is-layout-flex wp-block-buttons-is-layout-flex\">\n<div class=\"wp-block-button\"><a class=\"wp-block-button__link wp-element-button\" href=\"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/gsma_events\/cybersecurity-awareness-month-webinar-ai-use-in-social-engineering\/\">Register<\/a><\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>Editor\u2019s Note: This blog was originally published as part of our ongoing research into AI security in telecom and written by Khalifa University\u2019s Dr. Lina Bariah. We\u2019re republishing and updating &nbsp;to coincide with our upcoming webinar, \u201cSocial Engineering the Algorithm: A Red Teamer\u2019s Guide to AI\u201d, taking place on October 29. The insights from our [&hellip;]<\/p>\n","protected":false},"author":55,"featured_media":13725,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_oasis_is_in_workflow":0,"_oasis_original":0,"_oasis_task_priority":"","footnotes":"[]"},"categories":[1505],"tags":[],"algolia_discover_type":[],"class_list":["post-13719","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-latest-news"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v24.3 (Yoast SEO v24.3) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Cybersecurity Awareness Month: Building Trust in Telecom: Red Team Insights into Telecom AI Security - Security<\/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:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/\" \/>\n<meta property=\"og:locale\" content=\"en_GB\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Cybersecurity Awareness Month: Building Trust in Telecom: Red Team Insights into Telecom AI Security - Security\" \/>\n<meta property=\"og:description\" content=\"Editor\u2019s Note: This blog was originally published as part of our ongoing research into AI security in telecom and written by Khalifa University\u2019s Dr. Lina Bariah. We\u2019re republishing and updating &nbsp;to coincide with our upcoming webinar, \u201cSocial Engineering the Algorithm: A Red Teamer\u2019s Guide to AI\u201d, taking place on October 29. The insights from our [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/\" \/>\n<meta property=\"og:site_name\" content=\"Security\" \/>\n<meta property=\"article:published_time\" content=\"2025-10-24T13:49:49+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-10-27T14:23:22+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-content\/uploads\/2025\/10\/GettyImages-2218398165.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"683\" \/>\n\t<meta property=\"og:image:height\" content=\"512\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"phau@gsma.com\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"phau@gsma.com\" \/>\n\t<meta name=\"twitter:label2\" content=\"Estimated reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutes\" \/>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Cybersecurity Awareness Month: Building Trust in Telecom: Red Team Insights into Telecom AI Security - Security","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:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/","og_locale":"en_GB","og_type":"article","og_title":"Cybersecurity Awareness Month: Building Trust in Telecom: Red Team Insights into Telecom AI Security - Security","og_description":"Editor\u2019s Note: This blog was originally published as part of our ongoing research into AI security in telecom and written by Khalifa University\u2019s Dr. Lina Bariah. We\u2019re republishing and updating &nbsp;to coincide with our upcoming webinar, \u201cSocial Engineering the Algorithm: A Red Teamer\u2019s Guide to AI\u201d, taking place on October 29. The insights from our [&hellip;]","og_url":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/","og_site_name":"Security","article_published_time":"2025-10-24T13:49:49+00:00","article_modified_time":"2025-10-27T14:23:22+00:00","og_image":[{"width":683,"height":512,"url":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-content\/uploads\/2025\/10\/GettyImages-2218398165.jpg","type":"image\/jpeg"}],"author":"phau@gsma.com","twitter_card":"summary_large_image","twitter_misc":{"Written by":"phau@gsma.com","Estimated reading time":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/","url":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/","name":"Cybersecurity Awareness Month: Building Trust in Telecom: Red Team Insights into Telecom AI Security - Security","isPartOf":{"@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/#primaryimage"},"image":{"@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/#primaryimage"},"thumbnailUrl":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-content\/uploads\/2025\/10\/GettyImages-2218398165.jpg","datePublished":"2025-10-24T13:49:49+00:00","dateModified":"2025-10-27T14:23:22+00:00","author":{"@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/#\/schema\/person\/d5f2edfe4e539b3d59776c30df030bbe"},"inLanguage":"en-GB","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/"]}]},{"@type":"ImageObject","inLanguage":"en-GB","@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/latest-news\/building-trust-in-telecom-red-team-insights-into-telecom-ai-security\/#primaryimage","url":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-content\/uploads\/2025\/10\/GettyImages-2218398165.jpg","contentUrl":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-content\/uploads\/2025\/10\/GettyImages-2218398165.jpg","width":683,"height":512,"caption":"A hand holding a smartphone glows with red and blue light inside a bright, white, circular ring on a blue background, suggesting technology or digital interaction. The ring gives the scene a futuristic, high-tech feel."},{"@type":"WebSite","@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/#website","url":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/","name":"Security","description":"GSMA Security","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-GB"},{"@type":"Person","@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/#\/schema\/person\/d5f2edfe4e539b3d59776c30df030bbe","name":"phau@gsma.com","image":{"@type":"ImageObject","inLanguage":"en-GB","@id":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/5ecee0da5df233cb3e8fbbe640a36de3?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/5ecee0da5df233cb3e8fbbe640a36de3?s=96&d=mm&r=g","caption":"phau@gsma.com"}}]}},"featured_image_url":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-content\/uploads\/2025\/10\/GettyImages-2218398165.jpg","_links":{"self":[{"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/posts\/13719","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/users\/55"}],"replies":[{"embeddable":true,"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/comments?post=13719"}],"version-history":[{"count":3,"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/posts\/13719\/revisions"}],"predecessor-version":[{"id":13728,"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/posts\/13719\/revisions\/13728"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/media\/13725"}],"wp:attachment":[{"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/media?parent=13719"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/categories?post=13719"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/tags?post=13719"},{"taxonomy":"algolia_discover_type","embeddable":true,"href":"https:\/\/www.gsma.com\/solutions-and-impact\/technologies\/security\/wp-json\/wp\/v2\/algolia_discover_type?post=13719"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}