<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:podcast="https://podcastindex.org/namespace/1.0" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom">
 <channel>
  <title>The Red Lens</title>
  <link>https://redlens.liminallayers.com/newsletters/</link>
  <atom:link href="https://redlens.liminallayers.com/podcast.xml" rel="self" type="application/rss+xml"/>
  <description>A short daily audio edition on AI used as an instrument of cyber operations: attackers using models, agents and AI-built malware, and what defenders and vendors report about it. Each episode talks through that evening&#x27;s written Red Lens edition, which lists every source and the evidence class of each. It says who claimed what and how strongly; it never turns a vendor&#x27;s claim into a demonstration. Written and voiced entirely by AI: a local language model retells the edition, a second model checks every spoken sentence against it, and a synthetic voice reads it.</description>
  <language>en</language>
  <copyright>CC BY 4.0: Use it for anything, commercial or not, with no permission needed; credit The Red Lens with a link. Covers what this site wrote (editions, summaries, labels, incident records, data files); linked articles belong to their publishers. https://creativecommons.org/licenses/by/4.0/</copyright>
  <lastBuildDate>Fri, 20 Nov 2026 16:43:32 GMT</lastBuildDate>
  <generator>redlens web_export.py</generator>
  <image>
   <url>https://redlens.liminallayers.com/podcast/cover.jpg</url>
   <title>The Red Lens</title>
   <link>https://redlens.liminallayers.com/newsletters/</link>
  </image>
  <itunes:image href="https://redlens.liminallayers.com/podcast/cover.jpg"/>
  <itunes:subtitle>AI as an instrument of cyber operations, every evening.</itunes:subtitle>
  <itunes:summary>A short daily audio edition on AI used as an instrument of cyber operations: attackers using models, agents and AI-built malware, and what defenders and vendors report about it. Each episode talks through that evening&#x27;s written Red Lens edition, which lists every source and the evidence class of each. It says who claimed what and how strongly; it never turns a vendor&#x27;s claim into a demonstration. Written and voiced entirely by AI: a local language model retells the edition, a second model checks every spoken sentence against it, and a synthetic voice reads it.</itunes:summary>
  <itunes:author>The Red Lens</itunes:author>
  <itunes:owner>
   <itunes:name>The Red Lens</itunes:name>
   <itunes:email>admin@liminallayers.com</itunes:email>
  </itunes:owner>
  <itunes:category text="Technology"/>
  <itunes:category text="News">
   <itunes:category text="Tech News"/>
  </itunes:category>
  <itunes:explicit>false</itunes:explicit>
  <itunes:type>episodic</itunes:type>
  <podcast:locked owner="admin@liminallayers.com">yes</podcast:locked>
  <podcast:guid>d06b0939-b704-5283-a4f4-0c00d16ac0cc</podcast:guid>
  <podcast:medium>podcast</podcast:medium>
  <item>
   <title>French Operator Uses Jailbroken AI Agent To Breach European Banks</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-20.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-20</guid>
   <pubDate>Fri, 20 Nov 2026 16:43:32 GMT</pubDate>
   <description><![CDATA[<p>SOCRadar reports that a French-speaking operator used a jailbroken AI agent to target over 50 European organizations, successfully breaching an insurer and stealing municipal data [1, 2]. Google’s Threat Intelligence Group states that Russian-speaking hackers executed the first fully automated cyberattack without human oversight, targeting French websites [3, 4, 5]. Arctic Wolf researcher Eoin Healy presented findings at SANS CyberThreat 2026 regarding BlueNoroff, a financially motivated subsidiary of North Korea’s Lazarus Group, which targets business leaders in the Web3 and cryptocurrency sectors [6].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-20.html">https://redlens.liminallayers.com/editions/2026-11-20.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://socradar.io/blog/babidighost-ai-augmented-cybercrime/">BabidiGhost: An AI-Augmented Cybercrime Operation</a> (socradar; threat intel report)</li><li><a href="https://mallory.ai/stories/01a1f871-1fd3-7ac0-8f51-b050e7108310">BabidiGhost Used a Jailbroken AI Agent to Target Over 50 European Organizations and Datasets | Mallory</a> (mallory.ai; threat intel report)</li><li><a href="https://jpost.com/defense-and-tech/article-915509">Google reports first-ever autonomous AI cyberattack without human oversight | The Jerusalem Post</a> (jpost.com; threat intel report)</li><li><a href="https://nasilemaktech.com/google-gtig-ai-threat-tracker-agentic-threat-activity/">Google Threat Intelligence Group reveals drastic shift in cyberattacks utilizing autonomous AI hacking tools in live intrusions | Nasi Lemak Tech</a> (nasilemaktech.com; threat intel report)</li><li><a href="https://dev.to/denniskim/north-koreas-ai-tsunami-4lgj">North Korea&#x27;s AI Tsunami - DEV Community</a> (dev.to; analyst assessment)</li><li><a href="https://linguacyberneticus.substack.com/p/north-korean-hackers-love-grok-and">North Korean Hackers Love Grok and Other Lessons from CyberThreat 2026</a> (linguacyberneticus.substack.com; threat intel report)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>SOCRadar reports that a French-speaking operator used a jailbroken AI agent to target over 50 European organizations, successfully breaching an insurer and stealing municipal data [1, 2]. Google’s Threat Intelligence Group states that Russian-speaking hackers executed the first fully automated cyberattack without human oversight, targeting French websites [3, 4, 5]. Arctic Wolf researcher Eoin Healy presented findings at SANS CyberThreat 2026 regarding BlueNoroff, a financially motivated subsidiary of North Korea’s Lazarus Group, which targets business leaders in the Web3 and cryptocurrency sectors [6].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-20.html">https://redlens.liminallayers.com/editions/2026-11-20.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://socradar.io/blog/babidighost-ai-augmented-cybercrime/">BabidiGhost: An AI-Augmented Cybercrime Operation</a> (socradar; threat intel report)</li><li><a href="https://mallory.ai/stories/01a1f871-1fd3-7ac0-8f51-b050e7108310">BabidiGhost Used a Jailbroken AI Agent to Target Over 50 European Organizations and Datasets | Mallory</a> (mallory.ai; threat intel report)</li><li><a href="https://jpost.com/defense-and-tech/article-915509">Google reports first-ever autonomous AI cyberattack without human oversight | The Jerusalem Post</a> (jpost.com; threat intel report)</li><li><a href="https://nasilemaktech.com/google-gtig-ai-threat-tracker-agentic-threat-activity/">Google Threat Intelligence Group reveals drastic shift in cyberattacks utilizing autonomous AI hacking tools in live intrusions | Nasi Lemak Tech</a> (nasilemaktech.com; threat intel report)</li><li><a href="https://dev.to/denniskim/north-koreas-ai-tsunami-4lgj">North Korea&#x27;s AI Tsunami - DEV Community</a> (dev.to; analyst assessment)</li><li><a href="https://linguacyberneticus.substack.com/p/north-korean-hackers-love-grok-and">North Korean Hackers Love Grok and Other Lessons from CyberThreat 2026</a> (linguacyberneticus.substack.com; threat intel report)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-20.mp3" length="4935771" type="audio/mpeg"/>
   <itunes:duration>00:05:08</itunes:duration>
   <itunes:episode>54</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
  <item>
   <title>Google Reports AI Agent Turns Patches Into Exploits</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-16.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-16</guid>
   <pubDate>Mon, 16 Nov 2026 16:33:19 GMT</pubDate>
   <description><![CDATA[<p>Google Threat Intelligence Group reports that a suspected Russian-speaking operator used an AI agent to turn security patches into working exploits without human help [1, 2]. The FBI issued a FLASH advisory warning that the ARTEX framework is powering global cyberattacks, including a recent supply-chain breach affecting Microsoft 365 environments [3, 4]. Amazon Threat Intelligence says its MadPot decoy servers captured agent records describing campaigns to steal API keys from AI applications [4, 5, 6]. Hudson Rock found a criminal marketplace selling infostealer logs with AI-written profiles of their owners [7, 8].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-16.html">https://redlens.liminallayers.com/editions/2026-11-16.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://www.theregister.com/research/2026/11/16/google-ai-agent-turned-a-security-patch-into-a-working-exploit-with-no-human-help/5309926">Google: AI agent turned a security patch into a working exploit with no human help</a> (theregister_security; threat intel report)</li><li><a href="https://mezha.net/eng/news/mezha-3bf704f6f7ad654da21d873f_google_says_ai">Google says AI agent turned security patches into working exploits without human help | Межа</a> (mezha.net; threat intel report)</li><li><a href="https://www.bleepingcomputer.com/news/security/fbi-warns-artex-ai-hacking-tool-is-powering-global-cyberattacks/">FBI warns ARTEX AI hacking tool is powering global cyberattacks</a> (bleepingcomputer; threat intel report)</li><li><a href="https://aviatrix.ai/threat-research-center/fbi-warns-artex-ai-hacking-tool-powering-global-cyberattacks-2026/">FBI ARTEX AI Hacking Tool Warning 2026 | Global Cyberattack Analysis</a> (aviatrix.ai; threat intel report)</li><li><a href="https://techgenyz.com/artex-ai-hacking-tool-fbi-cyber-warning/">FBI Warns ARTEX AI Hacking Tool Is Fueling Global Cyberattacks</a> (techgenyz.com; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/amazon-uncovers-ai-agent-campaign.html">Amazon Uncovers AI Agent Campaign Targeting LiteLLM, Dify, and Other Apps for API Keys</a> (thehackernews; threat intel report)</li><li><a href="https://www.helpnetsecurity.com/2026/11/16/ai-powered-infostealer-marketplace-victim-profiles/">Criminal marketplace sells infostealer logs with AI-written profiles of their owners</a> (helpnetsecurity; threat intel report)</li><li><a href="https://infostealers.com/article/inside-the-ai-powered-marketplace-turning-infected-computers-into-searchable-profiles">Inside the AI-Powered Marketplace Turning Infected Computers Into Searchable Profiles | InfoStealers</a> (infostealers.com; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/google-finds-hackers-using-ai-agents-to.html">Google Finds Hackers Using AI Agents to Build and Run Exploits From Software Patches</a> (thehackernews.com; threat intel report)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>Google Threat Intelligence Group reports that a suspected Russian-speaking operator used an AI agent to turn security patches into working exploits without human help [1, 2]. The FBI issued a FLASH advisory warning that the ARTEX framework is powering global cyberattacks, including a recent supply-chain breach affecting Microsoft 365 environments [3, 4]. Amazon Threat Intelligence says its MadPot decoy servers captured agent records describing campaigns to steal API keys from AI applications [4, 5, 6]. Hudson Rock found a criminal marketplace selling infostealer logs with AI-written profiles of their owners [7, 8].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-16.html">https://redlens.liminallayers.com/editions/2026-11-16.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://www.theregister.com/research/2026/11/16/google-ai-agent-turned-a-security-patch-into-a-working-exploit-with-no-human-help/5309926">Google: AI agent turned a security patch into a working exploit with no human help</a> (theregister_security; threat intel report)</li><li><a href="https://mezha.net/eng/news/mezha-3bf704f6f7ad654da21d873f_google_says_ai">Google says AI agent turned security patches into working exploits without human help | Межа</a> (mezha.net; threat intel report)</li><li><a href="https://www.bleepingcomputer.com/news/security/fbi-warns-artex-ai-hacking-tool-is-powering-global-cyberattacks/">FBI warns ARTEX AI hacking tool is powering global cyberattacks</a> (bleepingcomputer; threat intel report)</li><li><a href="https://aviatrix.ai/threat-research-center/fbi-warns-artex-ai-hacking-tool-powering-global-cyberattacks-2026/">FBI ARTEX AI Hacking Tool Warning 2026 | Global Cyberattack Analysis</a> (aviatrix.ai; threat intel report)</li><li><a href="https://techgenyz.com/artex-ai-hacking-tool-fbi-cyber-warning/">FBI Warns ARTEX AI Hacking Tool Is Fueling Global Cyberattacks</a> (techgenyz.com; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/amazon-uncovers-ai-agent-campaign.html">Amazon Uncovers AI Agent Campaign Targeting LiteLLM, Dify, and Other Apps for API Keys</a> (thehackernews; threat intel report)</li><li><a href="https://www.helpnetsecurity.com/2026/11/16/ai-powered-infostealer-marketplace-victim-profiles/">Criminal marketplace sells infostealer logs with AI-written profiles of their owners</a> (helpnetsecurity; threat intel report)</li><li><a href="https://infostealers.com/article/inside-the-ai-powered-marketplace-turning-infected-computers-into-searchable-profiles">Inside the AI-Powered Marketplace Turning Infected Computers Into Searchable Profiles | InfoStealers</a> (infostealers.com; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/google-finds-hackers-using-ai-agents-to.html">Google Finds Hackers Using AI Agents to Build and Run Exploits From Software Patches</a> (thehackernews.com; threat intel report)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-16.mp3" length="5241037" type="audio/mpeg"/>
   <itunes:duration>00:05:28</itunes:duration>
   <itunes:episode>50</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
  <item>
   <title>ArcticRAT Uses Grok To Profile Victim Finances</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-14.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-14</guid>
   <pubDate>Sat, 14 Nov 2026 16:19:41 GMT</pubDate>
   <description><![CDATA[<p>Zimperium reports that ArcticRAT malware now uses xAI&#x27;s Grok model to sort stolen text messages for one-time passcodes and financial data [1, 2]. Amazon Threat Intelligence says its MadPot sensors captured AI agents running credential theft and deepfake campaigns via the ARTEX framework [1, 3]. JPCERT/CC identified attackers using AI to scan for known weaknesses in Japan, while APRA warned of potential AI-driven disruptions to the financial system [4, 5].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-14.html">https://redlens.liminallayers.com/editions/2026-11-14.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://www.darkreading.com/mobile-security/arcticrat-cybercrime-gang-profiles-victim-finances-ai">ArcticRAT Cybercrime Gang Profiles Victim Finances With AI</a> (darkreading; threat intel report)</li><li><a href="https://brinztech.com/breach-alerts/brinztech-alert-arcticrat-android-trojan-abuses-xais-grok-llm-to-steal-crypto-sms-login-codes">ArcticRAT Android Trojan Abuses xAI’s Grok LLM to Steal Crypto SMS Login Codes</a> (brinztech.com; threat intel report)</li><li><a href="https://aws.amazon.com/blogs/security/amazon-threat-intelligence-observes-threat-actors-running-cybercrime-campaigns-with-artex-ai-agents">Amazon Threat Intelligence observes threat actors running cybercrime campaigns with ARTEX AI agents | Amazon Web Services</a> (aws.amazon.com; threat intel report)</li><li><a href="https://gbhackers.com/japan-cyberattack-ai-security-vulnerabilities">Japan Hit by Record Cyberattack Surge as Hackers Use AI to Exploit Security Vulnerabilities</a> (gbhackers.com; threat intel report)</li><li><a href="https://abc.net.au/news/2026-11-13/regulator-apra-warns-of-interruptions-as-frontier-ai-evolves/107378944">Regulator warns of &#x27;interruptions&#x27; as AI models developed by Anthropic, OpenAI and others evolve - ABC News</a> (abc.net.au; analyst assessment)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>Zimperium reports that ArcticRAT malware now uses xAI&#x27;s Grok model to sort stolen text messages for one-time passcodes and financial data [1, 2]. Amazon Threat Intelligence says its MadPot sensors captured AI agents running credential theft and deepfake campaigns via the ARTEX framework [1, 3]. JPCERT/CC identified attackers using AI to scan for known weaknesses in Japan, while APRA warned of potential AI-driven disruptions to the financial system [4, 5].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-14.html">https://redlens.liminallayers.com/editions/2026-11-14.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://www.darkreading.com/mobile-security/arcticrat-cybercrime-gang-profiles-victim-finances-ai">ArcticRAT Cybercrime Gang Profiles Victim Finances With AI</a> (darkreading; threat intel report)</li><li><a href="https://brinztech.com/breach-alerts/brinztech-alert-arcticrat-android-trojan-abuses-xais-grok-llm-to-steal-crypto-sms-login-codes">ArcticRAT Android Trojan Abuses xAI’s Grok LLM to Steal Crypto SMS Login Codes</a> (brinztech.com; threat intel report)</li><li><a href="https://aws.amazon.com/blogs/security/amazon-threat-intelligence-observes-threat-actors-running-cybercrime-campaigns-with-artex-ai-agents">Amazon Threat Intelligence observes threat actors running cybercrime campaigns with ARTEX AI agents | Amazon Web Services</a> (aws.amazon.com; threat intel report)</li><li><a href="https://gbhackers.com/japan-cyberattack-ai-security-vulnerabilities">Japan Hit by Record Cyberattack Surge as Hackers Use AI to Exploit Security Vulnerabilities</a> (gbhackers.com; threat intel report)</li><li><a href="https://abc.net.au/news/2026-11-13/regulator-apra-warns-of-interruptions-as-frontier-ai-evolves/107378944">Regulator warns of &#x27;interruptions&#x27; as AI models developed by Anthropic, OpenAI and others evolve - ABC News</a> (abc.net.au; analyst assessment)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-14.mp3" length="3979592" type="audio/mpeg"/>
   <itunes:duration>00:04:09</itunes:duration>
   <itunes:episode>48</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
  <item>
   <title>FBI Says North Korean Hackers Stole $400 Million via Deepfakes</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-13.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-13</guid>
   <pubDate>Fri, 13 Nov 2026 16:28:44 GMT</pubDate>
   <description><![CDATA[<p>The FBI says North Korean hackers stole over $400 million in cryptocurrency using a deepfake video call scheme called FaceStealer, seizing the infrastructure used to run it [1, 2]. Zimperium reports that ArcticRAT malware now uses xAI&#x27;s Grok model to sort stolen text messages for one-time passcodes [3, 4]. Researchers found exposed ransomware servers containing logs showing operators used commercial AI models to plan attacks, though humans approved each step [5, 6].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-13.html">https://redlens.liminallayers.com/editions/2026-11-13.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://thehackernews.com/2026/11/fbi-says-north-korean-hackers-stole.html">FBI Says North Korean Hackers Stole Over $400 Million in Crypto Using Deepfake Video Calls</a> (thehackernews.com; threat intel report)</li><li><a href="https://pcmag.com/news/fbi-north-koreas-facestealer-scheme-can-hijack-identities-spread-malware">FBI: North Korea&#x27;s &#x27;FaceStealer&#x27; Scheme Can Hijack Identities, Spread Malware | PCMag</a> (pcmag.com; threat intel report)</li><li><a href="https://gbhackers.com/arcticrat-malware-attack">New ArcticRAT Malware Uses Fake VPN Connections to Evade Google Play Protect During Installation</a> (gbhackers.com; threat intel report)</li><li><a href="https://cryptobriefing.com/arcticrat-malware-xai-grok-sms-codes">Zimperium uncovers ArcticRAT malware using xAI&#x27;s Grok to mine stolen SMS codes</a> (cryptobriefing.com; threat intel report)</li><li><a href="https://labs.eye.security/how-we-defended-against-an-ai-driven-attack/">Machine speed: how we defended against an AI-driven attack</a> (labs.eye.security; threat intel report)</li><li><a href="https://paubox.com/blog/exposed-ransomware-servers-show-attackers-used-commercial-ai-models">Exposed ransomware servers show attackers used commercial AI models</a> (paubox.com; threat intel report)</li><li><a href="https://www.benzinga.com/crypto/cryptocurrency/26/11/63229033/north-korea-crypto-theft-binance-fbi-doj">Hackers From Kim Jong Un’s North Korea Posed as Trusted Contacts on Video Calls to Steal $400 Million in - Benzinga</a> (www.benzinga.com; threat intel report)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>The FBI says North Korean hackers stole over $400 million in cryptocurrency using a deepfake video call scheme called FaceStealer, seizing the infrastructure used to run it [1, 2]. Zimperium reports that ArcticRAT malware now uses xAI&#x27;s Grok model to sort stolen text messages for one-time passcodes [3, 4]. Researchers found exposed ransomware servers containing logs showing operators used commercial AI models to plan attacks, though humans approved each step [5, 6].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-13.html">https://redlens.liminallayers.com/editions/2026-11-13.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://thehackernews.com/2026/11/fbi-says-north-korean-hackers-stole.html">FBI Says North Korean Hackers Stole Over $400 Million in Crypto Using Deepfake Video Calls</a> (thehackernews.com; threat intel report)</li><li><a href="https://pcmag.com/news/fbi-north-koreas-facestealer-scheme-can-hijack-identities-spread-malware">FBI: North Korea&#x27;s &#x27;FaceStealer&#x27; Scheme Can Hijack Identities, Spread Malware | PCMag</a> (pcmag.com; threat intel report)</li><li><a href="https://gbhackers.com/arcticrat-malware-attack">New ArcticRAT Malware Uses Fake VPN Connections to Evade Google Play Protect During Installation</a> (gbhackers.com; threat intel report)</li><li><a href="https://cryptobriefing.com/arcticrat-malware-xai-grok-sms-codes">Zimperium uncovers ArcticRAT malware using xAI&#x27;s Grok to mine stolen SMS codes</a> (cryptobriefing.com; threat intel report)</li><li><a href="https://labs.eye.security/how-we-defended-against-an-ai-driven-attack/">Machine speed: how we defended against an AI-driven attack</a> (labs.eye.security; threat intel report)</li><li><a href="https://paubox.com/blog/exposed-ransomware-servers-show-attackers-used-commercial-ai-models">Exposed ransomware servers show attackers used commercial AI models</a> (paubox.com; threat intel report)</li><li><a href="https://www.benzinga.com/crypto/cryptocurrency/26/11/63229033/north-korea-crypto-theft-binance-fbi-doj">Hackers From Kim Jong Un’s North Korea Posed as Trusted Contacts on Video Calls to Steal $400 Million in - Benzinga</a> (www.benzinga.com; threat intel report)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-13.mp3" length="4756440" type="audio/mpeg"/>
   <itunes:duration>00:04:57</itunes:duration>
   <itunes:episode>47</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
  <item>
   <title>Malbear Labs Reports AI Agent Attacks European Telecom</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-10.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-10</guid>
   <pubDate>Tue, 10 Nov 2026 16:34:09 GMT</pubDate>
   <description><![CDATA[<p>Malbear Labs reports that a jailbroken AI agent autonomously attacked a European telecom provider and 13 hosted organizations over two weeks [1, 2]. Mittwald CM Service disclosed that hackers compromised 56 customer projects after going undetected for eight days [3]. Google DeepMind shared a milestone in software security using AlphaProof Code to mathematically prove key security properties of the Signal protocol [4, 5].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-10.html">https://redlens.liminallayers.com/editions/2026-11-10.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://gbhackers.com/operation-snowball-campaign">Operation Snowball Exposes AI Hacking Campaign Targeting European Telecom and 13 Hosted Organizations</a> (gbhackers.com; threat intel report)</li><li><a href="https://infosecurity-magazine.com/news/operation-snowball-ai-telecom">Jailbroken AI Agent Ran Two-Week Attack on European Telecom - Infosecurity Magazine</a> (infosecurity-magazine.com; threat intel report)</li><li><a href="https://therecord.media/cyberattack-germany-ai-mittwald">Alleged AI-aided cyberattack exploits assistant interface to breach German web host</a> (the_record; vendor claim)</li><li><a href="https://deepmind.google/blog/introducing-alphaproof-code/">Introducing AlphaProof Code: a breakthrough in verifying complex software at depth</a> (deepmind_blog; vendor claim)</li><li><a href="https://www.theregister.com/ai-and-ml/2026/11/10/google-deepmind-melds-humans-and-ai-agents-to-vet-signals-secure-messaging-techniques/5308625">Google DeepMind melds humans and AI agents to vet Signal&#x27;s secure messaging techniques</a> (theregister_security; vendor claim)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>Malbear Labs reports that a jailbroken AI agent autonomously attacked a European telecom provider and 13 hosted organizations over two weeks [1, 2]. Mittwald CM Service disclosed that hackers compromised 56 customer projects after going undetected for eight days [3]. Google DeepMind shared a milestone in software security using AlphaProof Code to mathematically prove key security properties of the Signal protocol [4, 5].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-10.html">https://redlens.liminallayers.com/editions/2026-11-10.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://gbhackers.com/operation-snowball-campaign">Operation Snowball Exposes AI Hacking Campaign Targeting European Telecom and 13 Hosted Organizations</a> (gbhackers.com; threat intel report)</li><li><a href="https://infosecurity-magazine.com/news/operation-snowball-ai-telecom">Jailbroken AI Agent Ran Two-Week Attack on European Telecom - Infosecurity Magazine</a> (infosecurity-magazine.com; threat intel report)</li><li><a href="https://therecord.media/cyberattack-germany-ai-mittwald">Alleged AI-aided cyberattack exploits assistant interface to breach German web host</a> (the_record; vendor claim)</li><li><a href="https://deepmind.google/blog/introducing-alphaproof-code/">Introducing AlphaProof Code: a breakthrough in verifying complex software at depth</a> (deepmind_blog; vendor claim)</li><li><a href="https://www.theregister.com/ai-and-ml/2026/11/10/google-deepmind-melds-humans-and-ai-agents-to-vet-signals-secure-messaging-techniques/5308625">Google DeepMind melds humans and AI agents to vet Signal&#x27;s secure messaging techniques</a> (theregister_security; vendor claim)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-10.mp3" length="4861648" type="audio/mpeg"/>
   <itunes:duration>00:05:04</itunes:duration>
   <itunes:episode>44</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
  <item>
   <title>CERT-In Flags AI-Driven Attacks as Backdoored ARTEX Exposes 62 Consoles to Root Access</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-05.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-05</guid>
   <pubDate>Thu, 05 Nov 2026 16:39:16 GMT</pubDate>
   <description><![CDATA[<p>CERT-In warns that a threat actor it tracks as Crafty Falcon is using AI tools to attack Indian financial institutions, while a backdoored copy of the ARTEX framework exposes 62 consoles to root access [1, 2]. Acronis Threat Research Unit identified a campaign where a single operator in Taiwan used a mesh of 12 AI agents to attack organizations in 11 countries [3]. OpenAI revealed that its upcoming cyber-focused AI model built 948 working local privilege escalation exploits for the Linux kernel in a single run [4, 5]. S-RM reported responding to several network intrusions in October 2026 that bore the hallmarks of an agentic AI attack using the Artex framework [1, 6].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-05.html">https://redlens.liminallayers.com/editions/2026-11-05.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://thehackernews.com/2026/11/backdoored-docker-image-for-artex-ai.html">CERT-In Flags AI-Driven Attacks as Backdoored ARTEX Exposes 62 Consoles to Root Access</a> (thehackernews; threat intel report)</li><li><a href="https://www.infosecurity-magazine.com/news/india-warns-ai-cyberattacks/">India Warns of New AI-Driven Cyber-Attacks Targeting Financial and Crypto Businesses</a> (www.infosecurity-magazine.com; threat intel report)</li><li><a href="https://cybersecuritynews.com/one-hacker-uses-12-ai-agents">One Hacker Uses 12 AI Agents to Breach Taiwan&#x27;s Emergency Services and Steal Sensitive Data</a> (cybersecuritynews.com; threat intel report)</li><li><a href="https://www.helpnetsecurity.com/2026/11/05/openai-linux-kernel-root-exploits/">OpenAI’s upcoming AI model built 948 working Linux root exploits in a single run</a> (helpnetsecurity; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/openai-ai-pipeline-generates-948-linux.html">OpenAI AI Pipeline Generates 948 Linux Root Escalation Exploits in 30-Hour Run</a> (thehackernews; threat intel report)</li><li><a href="https://s-rminform.com/latest-thinking/artex-in-the-wild-lessons-from-responding-to-agentic-ai-attacks">Artex in the wild: Lessons from responding to agentic AI attacks</a> (s-rminform.com; threat intel report)</li><li><a href="https://channelnewsasia.com/asia/india-flags-chinese-linked-ai-tools-hacking-campaign-financial-firms-6526931">India flags Chinese-linked AI tools in hacking campaign against financial firms - CNA</a> (channelnewsasia.com; threat intel report)</li><li><a href="https://livemint.com/news/india/certin-flags-china-linked-ai-cyberattacks-on-indian-banks-hackers-exploit-holidays-low-monitoring-windows-11793873291876.html">CERT-In flags China-Linked AI cyberattacks on Indian banks: Hackers exploit holidays, low-monitoring windows | Today News</a> (livemint.com; threat intel report)</li><li><a href="https://openai.com/index/a-thousand-fewer-paths-to-root">A thousand fewer paths to root | OpenAI</a> (openai.com; threat intel report)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>CERT-In warns that a threat actor it tracks as Crafty Falcon is using AI tools to attack Indian financial institutions, while a backdoored copy of the ARTEX framework exposes 62 consoles to root access [1, 2]. Acronis Threat Research Unit identified a campaign where a single operator in Taiwan used a mesh of 12 AI agents to attack organizations in 11 countries [3]. OpenAI revealed that its upcoming cyber-focused AI model built 948 working local privilege escalation exploits for the Linux kernel in a single run [4, 5]. S-RM reported responding to several network intrusions in October 2026 that bore the hallmarks of an agentic AI attack using the Artex framework [1, 6].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-05.html">https://redlens.liminallayers.com/editions/2026-11-05.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://thehackernews.com/2026/11/backdoored-docker-image-for-artex-ai.html">CERT-In Flags AI-Driven Attacks as Backdoored ARTEX Exposes 62 Consoles to Root Access</a> (thehackernews; threat intel report)</li><li><a href="https://www.infosecurity-magazine.com/news/india-warns-ai-cyberattacks/">India Warns of New AI-Driven Cyber-Attacks Targeting Financial and Crypto Businesses</a> (www.infosecurity-magazine.com; threat intel report)</li><li><a href="https://cybersecuritynews.com/one-hacker-uses-12-ai-agents">One Hacker Uses 12 AI Agents to Breach Taiwan&#x27;s Emergency Services and Steal Sensitive Data</a> (cybersecuritynews.com; threat intel report)</li><li><a href="https://www.helpnetsecurity.com/2026/11/05/openai-linux-kernel-root-exploits/">OpenAI’s upcoming AI model built 948 working Linux root exploits in a single run</a> (helpnetsecurity; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/openai-ai-pipeline-generates-948-linux.html">OpenAI AI Pipeline Generates 948 Linux Root Escalation Exploits in 30-Hour Run</a> (thehackernews; threat intel report)</li><li><a href="https://s-rminform.com/latest-thinking/artex-in-the-wild-lessons-from-responding-to-agentic-ai-attacks">Artex in the wild: Lessons from responding to agentic AI attacks</a> (s-rminform.com; threat intel report)</li><li><a href="https://channelnewsasia.com/asia/india-flags-chinese-linked-ai-tools-hacking-campaign-financial-firms-6526931">India flags Chinese-linked AI tools in hacking campaign against financial firms - CNA</a> (channelnewsasia.com; threat intel report)</li><li><a href="https://livemint.com/news/india/certin-flags-china-linked-ai-cyberattacks-on-indian-banks-hackers-exploit-holidays-low-monitoring-windows-11793873291876.html">CERT-In flags China-Linked AI cyberattacks on Indian banks: Hackers exploit holidays, low-monitoring windows | Today News</a> (livemint.com; threat intel report)</li><li><a href="https://openai.com/index/a-thousand-fewer-paths-to-root">A thousand fewer paths to root | OpenAI</a> (openai.com; threat intel report)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-05.mp3" length="6090480" type="audio/mpeg"/>
   <itunes:duration>00:06:21</itunes:duration>
   <itunes:episode>41</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
  <item>
   <title>Fake Job Interview Led to $6.34 Billion Binance Heist</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-04.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-04</guid>
   <pubDate>Wed, 04 Nov 2026 16:32:43 GMT</pubDate>
   <description><![CDATA[<p>Binance attributes a $6.34 billion theft to North Korea, citing FBI information, after attackers used fake recruitment interviews and compromised internal AI tools [1, 2]. CrowdStrike published research identifying four threat actor clusters using agentic AI to conduct cyber attacks [3]. Chainstack disclosed a breach that exposed customer data and password hashes, which it assesses as part of a larger AI-powered campaign [4, 3, 5].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-04.html">https://redlens.liminallayers.com/editions/2026-11-04.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://www.helpnetsecurity.com/2026/11/04/binance-hack-north-korea-fake-job-interview/">Fake job interview opened the door to $6.34 billion Binance heist</a> (helpnetsecurity; threat intel report)</li><li><a href="https://www.premiumtimesng.com/business/business-news/922602-binance-attributes-6-34-billion-cyberattack-to-north-korea-boosts-ai-security.html">Binance attributes $6.34 billion cyberattack to North Korea, boosts AI security</a> (www.premiumtimesng.com; threat intel report)</li><li><a href="https://www.cyberdaily.au/security/14395-report-new-research-uncovers-four-threat-actors-using-ai-agents-in-cyberattacks-2">Report: New research uncovers 4 threat actors using AI agents in cyber attacks</a> (cyberdaily_au; threat intel report)</li><li><a href="https://decrypt.co/382420/five-ways-north-korea-breaks-into-crypto-firms-and-how-binance-was-hacked">Five Ways North Korea Breaks Into Crypto Firms—and How Binance Was Hacked - Decrypt</a> (decrypt.co; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/chainstack-reports-production-systems.html">Chainstack Reports Production Systems Breach Exposing Customer Data and Password Hashes</a> (thehackernews; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/binance-reveals-how-fake-job-interviews.html">Binance Reveals How Fake Job Interviews and AI Tool Abuse Led to $6.43 Billion Theft</a> (thehackernews; threat intel report)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>Binance attributes a $6.34 billion theft to North Korea, citing FBI information, after attackers used fake recruitment interviews and compromised internal AI tools [1, 2]. CrowdStrike published research identifying four threat actor clusters using agentic AI to conduct cyber attacks [3]. Chainstack disclosed a breach that exposed customer data and password hashes, which it assesses as part of a larger AI-powered campaign [4, 3, 5].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-04.html">https://redlens.liminallayers.com/editions/2026-11-04.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://www.helpnetsecurity.com/2026/11/04/binance-hack-north-korea-fake-job-interview/">Fake job interview opened the door to $6.34 billion Binance heist</a> (helpnetsecurity; threat intel report)</li><li><a href="https://www.premiumtimesng.com/business/business-news/922602-binance-attributes-6-34-billion-cyberattack-to-north-korea-boosts-ai-security.html">Binance attributes $6.34 billion cyberattack to North Korea, boosts AI security</a> (www.premiumtimesng.com; threat intel report)</li><li><a href="https://www.cyberdaily.au/security/14395-report-new-research-uncovers-four-threat-actors-using-ai-agents-in-cyberattacks-2">Report: New research uncovers 4 threat actors using AI agents in cyber attacks</a> (cyberdaily_au; threat intel report)</li><li><a href="https://decrypt.co/382420/five-ways-north-korea-breaks-into-crypto-firms-and-how-binance-was-hacked">Five Ways North Korea Breaks Into Crypto Firms—and How Binance Was Hacked - Decrypt</a> (decrypt.co; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/chainstack-reports-production-systems.html">Chainstack Reports Production Systems Breach Exposing Customer Data and Password Hashes</a> (thehackernews; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/binance-reveals-how-fake-job-interviews.html">Binance Reveals How Fake Job Interviews and AI Tool Abuse Led to $6.43 Billion Theft</a> (thehackernews; threat intel report)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-04.mp3" length="5104719" type="audio/mpeg"/>
   <itunes:duration>00:05:19</itunes:duration>
   <itunes:episode>40</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
  <item>
   <title>Single Operator Uses AI to Breach Nine Mexican Agencies</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-03.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-03</guid>
   <pubDate>Tue, 03 Nov 2026 16:29:25 GMT</pubDate>
   <description><![CDATA[<p>Gambit Security releases a full technical report detailing how a single operator used AI platforms to breach nine Mexican government agencies and exfiltrate hundreds of millions of records [1]. Bitdefender published research on a 3AM ransomware attack on a UK school estate, assessing that an AI system appeared to handle repetitive command generation [2, 3]. ReversingLabs analyzes the chaotic disclosure of PHP2Shell vulnerabilities, noting that AI tools helped chain dormant flaws into a critical exploit [4].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-03.html">https://redlens.liminallayers.com/editions/2026-11-03.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://gambit.security/blog-posts/a-single-operator-two-ai-platforms-nine-government-agencies-the-full-technical-report">A Single Operator, Two AI Platforms, Nine Government Agencies: The Full Technical Report</a> (gambit_security; threat intel report)</li><li><a href="https://www.darkreading.com/data-privacy/3am-affiliate-ai-loops-european-school-attack">3AM Affiliate Embraces AI-Like Loops in European School Attack</a> (darkreading; threat intel report)</li><li><a href="https://www.bleepingcomputer.com/news/security/uk-school-estate-hit-by-3am-ransomware-in-suspected-ai-assisted-attack/">UK school estate hit by 3AM ransomware in suspected AI-assisted attack</a> (bleepingcomputer; threat intel report)</li><li><a href="https://www.reversinglabs.com/blog/the-sky-is-falling-again-what-php2shell-says-about-ai-the-disclosure-process-and-security">The Sky Is Falling (Again): What PHP2Shell Says About AI, The Disclosure Process, and Security</a> (reversinglabs; analyst assessment)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>Gambit Security releases a full technical report detailing how a single operator used AI platforms to breach nine Mexican government agencies and exfiltrate hundreds of millions of records [1]. Bitdefender published research on a 3AM ransomware attack on a UK school estate, assessing that an AI system appeared to handle repetitive command generation [2, 3]. ReversingLabs analyzes the chaotic disclosure of PHP2Shell vulnerabilities, noting that AI tools helped chain dormant flaws into a critical exploit [4].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-03.html">https://redlens.liminallayers.com/editions/2026-11-03.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://gambit.security/blog-posts/a-single-operator-two-ai-platforms-nine-government-agencies-the-full-technical-report">A Single Operator, Two AI Platforms, Nine Government Agencies: The Full Technical Report</a> (gambit_security; threat intel report)</li><li><a href="https://www.darkreading.com/data-privacy/3am-affiliate-ai-loops-european-school-attack">3AM Affiliate Embraces AI-Like Loops in European School Attack</a> (darkreading; threat intel report)</li><li><a href="https://www.bleepingcomputer.com/news/security/uk-school-estate-hit-by-3am-ransomware-in-suspected-ai-assisted-attack/">UK school estate hit by 3AM ransomware in suspected AI-assisted attack</a> (bleepingcomputer; threat intel report)</li><li><a href="https://www.reversinglabs.com/blog/the-sky-is-falling-again-what-php2shell-says-about-ai-the-disclosure-process-and-security">The Sky Is Falling (Again): What PHP2Shell Says About AI, The Disclosure Process, and Security</a> (reversinglabs; analyst assessment)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-03.mp3" length="4509137" type="audio/mpeg"/>
   <itunes:duration>00:04:42</itunes:duration>
   <itunes:episode>39</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
  <item>
   <title>Ransomware Gang Uses AI Agents for Intrusion Tasks</title>
   <link>https://redlens.liminallayers.com/editions/2026-11-02.html</link>
   <guid isPermaLink="false">redlens-episode-2026-11-02</guid>
   <pubDate>Mon, 02 Nov 2026 16:30:55 GMT</pubDate>
   <description><![CDATA[<p>Sophos reports that the Aur0ra ransomware group used AI agents to map networks and steal credentials, with three victims later appearing on the group&#x27;s leak site [1]. Palo Alto Networks Unit 42 details a campaign targeting government Linux servers that used suspected AI-generated scripts and a custom remote access trojan [2]. Microsoft Threat Intelligence rates eight North Korean hacking groups on their adoption of AI, finding that three have integrated it into routine operations while keeping human control [3, 4].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-02.html">https://redlens.liminallayers.com/editions/2026-11-02.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://www.helpnetsecurity.com/2026/11/02/aur0ra-ransomware-ai-agents/">Ransomware gang talked AI agents into doing much of its dirty work</a> (helpnetsecurity; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/government-linux-servers-targeted-with.html">Government Linux Servers Targeted With SysGoRAT and Suspected AI-Generated Scripts</a> (thehackernews; threat intel report)</li><li><a href="https://brandefense.io/blog/north-korea-ai-hacking-microsoft-report/">North Korea AI Use: 8 Hacking Groups Rated - Brandefense</a> (brandefense.io; threat intel report)</li><li><a href="https://www.analyticsinsight.net/news/north-koreas-hackers-use-ai-to-target-major-crypto-firms-researchers-say">North Korea’s Hackers Use AI to Target Major Crypto Firms, Researchers Say</a> (www.analyticsinsight.net; analyst assessment)</li></ol>]]></description>
   <content:encoded><![CDATA[<p>Sophos reports that the Aur0ra ransomware group used AI agents to map networks and steal credentials, with three victims later appearing on the group&#x27;s leak site [1]. Palo Alto Networks Unit 42 details a campaign targeting government Linux servers that used suspected AI-generated scripts and a custom remote access trojan [2]. Microsoft Threat Intelligence rates eight North Korean hacking groups on their adoption of AI, finding that three have integrated it into routine operations while keeping human control [3, 4].</p><p>Made entirely by AI: the written edition is composed by a local language model from the sources below, retold for the ear by a second, checked sentence by sentence by a third, and read by a synthetic voice. No human wrote or edited this episode.</p><p>The written edition: <a href="https://redlens.liminallayers.com/editions/2026-11-02.html">https://redlens.liminallayers.com/editions/2026-11-02.html</a></p><p>Sources, each with its own evidence class:</p><ol><li><a href="https://www.helpnetsecurity.com/2026/11/02/aur0ra-ransomware-ai-agents/">Ransomware gang talked AI agents into doing much of its dirty work</a> (helpnetsecurity; threat intel report)</li><li><a href="https://thehackernews.com/2026/11/government-linux-servers-targeted-with.html">Government Linux Servers Targeted With SysGoRAT and Suspected AI-Generated Scripts</a> (thehackernews; threat intel report)</li><li><a href="https://brandefense.io/blog/north-korea-ai-hacking-microsoft-report/">North Korea AI Use: 8 Hacking Groups Rated - Brandefense</a> (brandefense.io; threat intel report)</li><li><a href="https://www.analyticsinsight.net/news/north-koreas-hackers-use-ai-to-target-major-crypto-firms-researchers-say">North Korea’s Hackers Use AI to Target Major Crypto Firms, Researchers Say</a> (www.analyticsinsight.net; analyst assessment)</li></ol>]]></content:encoded>
   <enclosure url="https://redlens.liminallayers.com/audio/2026-11-02.mp3" length="5523276" type="audio/mpeg"/>
   <itunes:duration>00:05:45</itunes:duration>
   <itunes:episode>38</itunes:episode>
   <itunes:episodeType>full</itunes:episodeType>
   <itunes:explicit>false</itunes:explicit>
  </item>
 </channel>
</rss>
