{"generated_at": "2026-07-14T03:17:59.974913Z", "slug": "security_now_audio", "source_id": "src_security_now_audio", "name": "Security Now (Audio)", "episode_count": 9, "avg_signal": 78.1, "median_signal": 77.2, "top_signal": 93.4, "latest_episode_at": "2026-07-08T02:19:20Z", "earliest_episode_at": "2026-05-13T02:36:18Z", "category_mode": "ai", "cover_image_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/07/862074/hero/SN1086_thumbnail.jpg", "rank_score": 167.71, "episodes": [{"episode_id": "ep_security_now_audio_a45389eae552", "episode_title": "SN 1086: The Apex Agentic Adversary - Visual Prompt Injection Strikes", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-07-08T02:19:20Z", "overall_score": 76.2, "score_breakdown": {"clarity": 52.0, "originality": 53.8, "hype_penalty": 9.0, "actionability": 100.0, "technical_depth": 100.0, "information_density": 100.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/07/862074/hero/SN1086_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1086", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1086/ARML1942480908.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1086", "verdict": "worth_your_time", "why_listen": "It goes beyond the title with direct discussion of it's, know, that's, including: Steve Gibson is here.", "summary": "Steve Gibson is here. Actually, we're going to talk about AI used for security."}, {"episode_id": "ep_security_now_audio_146a29db868d", "episode_title": "SN 1085: A SOTA State-Sponsored Campaign - AI's New Superpower: Loop Engineering", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-07-01T01:43:36Z", "overall_score": 84.4, "score_breakdown": {"clarity": 80.2, "originality": 51.8, "hype_penalty": 6.0, "actionability": 100.0, "technical_depth": 100.0, "information_density": 100.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/06/861994/hero/SN1085_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1085", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1085/ARML1669131272.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1085", "verdict": "worth_your_time", "why_listen": "It goes beyond the title with direct discussion of steve, gibson, it's, including: Time codes refer to the approximate times in the ad-free version of the show.", "summary": "Security Now 1085 transcript Primary Navigation Podcasts Club Blog Subscribe Sponsors More\u2026 Transcripts Security Now 1085 transcript Jun 30th 2026 Please be advised that this transcript is AI-generated and may not be word-for-word. Leo Laporte [00:00:00]: It's time for Security now."}, {"episode_id": "ep_security_now_audio_b400cd25b57e", "episode_title": "SN 1084: The Residential Proxy Threat - Malicious Proxies in Your Living Room", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-06-24T02:33:10Z", "overall_score": 82.3, "score_breakdown": {"clarity": 80.4, "originality": 56.1, "hype_penalty": 9.0, "actionability": 100.0, "technical_depth": 100.0, "information_density": 100.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/06/861905/hero/SN1084_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1084", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1084/ARML9361381102.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1084", "verdict": "worth_your_time", "why_listen": "It goes beyond the title with direct discussion of steve, gibson, know, including: Time codes refer to the approximate times in the ad-free version of the show.", "summary": "Security Now 1084 transcript Primary Navigation Podcasts Club Blog Subscribe Sponsors More\u2026 Transcripts Security Now 1084 transcript Jun 23rd 2026 Please be advised that this transcript is AI-generated and may not be word-for-word. Leo Laporte [00:00:00]: It's time for Security now."}, {"episode_id": "ep_security_now_audio_29e53fc89ec1", "episode_title": "SN 1083: Patch Tuesday \u00e0 la AI - Arch Linux Repo Under Siege", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-06-17T02:14:17Z", "overall_score": 68.8, "score_breakdown": {"clarity": 47.1, "originality": 51.8, "hype_penalty": 15.0, "actionability": 100.0, "technical_depth": 100.0, "information_density": 100.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/06/861797/hero/SN1083_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1083", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1083/ARML1367165950.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1083", "verdict": "worth_your_time", "why_listen": "It goes beyond the title with direct discussion of it's, know, like, including: Steve Gibson is here and he's raring to go.", "summary": "Steve Gibson is here and he's raring to go."}, {"episode_id": "ep_security_now_audio_5c2aab3349c1", "episode_title": "SN 1082: The Malicious Use of AI - Anthropic's Red Team Report", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-06-10T01:17:35Z", "overall_score": 77.2, "score_breakdown": {"clarity": 61.1, "originality": 49.8, "hype_penalty": 9.0, "actionability": 100.0, "technical_depth": 100.0, "information_density": 100.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/06/861713/hero/SN1082_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1082", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1082/ARML4767501078.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1082", "verdict": "worth_your_time", "why_listen": "It goes beyond the title with direct discussion of know, it's, which, including: Steve Gibson is here.", "summary": "Steve Gibson is here. It's appropriate because we're going to talk about how what Anthropic has learned from years of abuse of their AI models."}, {"episode_id": "ep_security_now_audio_7286135e8204", "episode_title": "SN 1081: AI Captured the Flag - Personal AI: Productivity Superpower or Privacy Threat?", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-06-03T03:08:50Z", "overall_score": 66.1, "score_breakdown": {"clarity": 47.0, "originality": 53.5, "hype_penalty": 18.0, "actionability": 100.0, "technical_depth": 100.0, "information_density": 100.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/06/861619/hero/SN1081_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1081", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1081/ARML3119278289.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1081", "verdict": "worth_your_time", "why_listen": "It goes beyond the title with direct discussion of know, it's, like, including: The end of a very popular hacker competition.", "summary": "The end of a very popular hacker competition."}, {"episode_id": "ep_security_now_audio_85a1b94021e8", "episode_title": "SN 1080: Vulnerability Debt Repayment - Will Mythos Change Cybersecurity Forever?", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-05-27T03:00:00Z", "overall_score": 76.2, "score_breakdown": {"clarity": 75.0, "originality": 94.0, "hype_penalty": 5.0, "actionability": 60.0, "technical_depth": 82.0, "information_density": 75.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/05/861521/hero/SN1080_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1080", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1080/ARML8286176864.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1080", "verdict": "must_listen", "why_listen": "Understand how AI models like Mythos are redefining vulnerability discovery and forcing real-time adaptation in cybersecurity defense strategies.", "summary": "AI models like Anthropic's Mythos are demonstrating unprecedented capability to find and exploit software vulnerabilities, forcing a fundamental shift in cybersecurity defense strategies. Cisco reports that Mythos has already changed their threat modeling and defensive posture, anticipating AI-powered attackers will become widespread. Mozilla's use of Mythos uncovered critical vulnerabilities in Firefox, suggesting AI-driven vulnerability discovery is not only real but already accelerating the repayment of long-accumulated vulnerability debt."}, {"episode_id": "ep_security_now_audio_1b4326505eed", "episode_title": "SN 1079: Daybreak and Codename MDASH - Microsoft's Edge Password Blunder", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-05-20T03:09:17Z", "overall_score": 78.4, "score_breakdown": {"clarity": 85.0, "originality": 95.0, "hype_penalty": 5.0, "actionability": 60.0, "technical_depth": 82.0, "information_density": 75.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/05/861427/hero/SN1079_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1079", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1079/ARML1459185120.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1079", "verdict": "must_listen", "why_listen": "Understand how AI is transforming cybersecurity from reactive patching to proactive, systematic vulnerability discovery at scale.", "summary": "Microsoft's Codename EM Dash, an AI-powered multi-agent system, discovered 16 critical vulnerabilities in Windows, including four remote code execution flaws, which were patched in the latest Patch Tuesday. The system uses over 100 specialized AI agents that collaborate, debate, and validate findings to reduce false positives and prove exploitability, marking a shift from reactive patching to proactive, AI-driven security. This approach, built on expertise from DARPA AI Cyber Challenge winners, sets a new benchmark for scalable, industry-wide software security."}, {"episode_id": "ep_security_now_audio_5293b806dce5", "episode_title": "SN 1078: DigiCert does it right - Hugging Face Under Fire", "podcast_name": "Security Now (Audio)", "podcast_slug": "security_now_audio", "source_id": "src_security_now_audio", "category": "ai", "publish_date": "2026-05-13T02:36:18Z", "overall_score": 93.4, "score_breakdown": {"clarity": 95.0, "originality": 95.0, "hype_penalty": 1.0, "actionability": 85.0, "technical_depth": 92.0, "information_density": 85.0}, "podcast_cover_url": "https://elroy.twit.tv/sites/default/files/styles/twit_thumb_720x405/public/images/episodes/2026/05/861331/hero/SN1078_thumbnail.jpg", "source_link": "https://twit.tv/shows/security-now/episodes/1078", "audio_url": "https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/megaphone/sn_1078/ARML5203491657.mp3", "listen_url": "https://twit.tv/shows/security-now/episodes/1078", "verdict": "must_listen", "why_listen": "Learn how a gold-standard incident response from DigiCert contrasted with Microsoft's overzealous automation, revealing critical lessons in trust management and digital certificate security.", "summary": "DigiCert's breach involved 27 fraudulent code signing certificates issued after attackers exploited a support portal vulnerability, but rapid revocation and transparent reporting limited damage. The company's full disclosure, immediate revocation from the original issue date, and systemic remediation set a new standard for incident response. In contrast, Microsoft Defender's overreaction\u2014removing DigiCert's root certificates from trust stores\u2014caused widespread disruption by breaking legitimate software trust."}], "category_breakdown": [{"category": "ai", "count": 9}]}