Security Unfiltered

Hosted by Joe South

Technology Business Careers NewsInterviews guests

Website RSS feed

Episodes

288

Latest episode

Oct 2026

Language

About the show

Cyber Security can be a difficult field to not only understand but to also navigate. Joe South is here to help with over a decade of experience across several domains of security. With this podcast I hope to help more people get into IT and Cyber Security as well as discussing modern day Cyber Security topics you may find in the daily news. Come join us as we learn and grow together!

Listen to episodes

60 recent

October 22, 2026Episode 28753 min

AI Agents Are Deleting Production. Cloud Security Has to Catch Up | Matt Lea

Matt Lea, founder of Schematical and an AWS consultant since 2010, joins Joe South on Security Unfiltered. From dropping out of college to crash a startup, to calling a CEO straight from his WHOIS record, Matt explains why he bet his career on infrastructure, and why vibe coding is creating a whole new class of breaches. They get into the OpenAI agent that escaped its sandbox, model distillation, stolen API keys, an AI agent that posted environment data to Postbin, and the Terraform change that wiped a production database and its backups. Plus AWS vs Azure vs GCP, why Azure MFA took 12 checks to verify, Cloud War Games for training incident responders under pressure, the "lone wolf programmer" problem, and where to focus if you're breaking into cloud and AI security now. Chapters 00:00 Seven months in the making (and a tornado) 01:40 Matt's path: Q Basic, a failed startup, AWS in 2010 04:03 Why infrastructure is the steady career 04:51 AI security is cloud security (for now) 05:44 Vibe coding and the mid-market security gap 06:39 "AI will take your job" vs reality 08:22 Organized crime goes cyber 12:28 Startups, scale and constant attacks 13:24 OpenAI's agent escaped the sandbox 18:02 Models are now worth stealing 19:09 Model distillation explained 23:35 Stolen API keys and vibe coding 27:00 AI agents gone wrong: Postbin and a deleted prod database 30:17 Why AWS (and when GCP) 32:03 Azure frustration vs AWS 34:37 GCP security by default 37:46 Twelve places to check MFA in Azure 39:35 Security spend vs runway 40:47 Cloud War Games: training under pressure 44:01 Tabletops, runbooks and the lone wolf programmer 48:47 Where to start learning cloud and AI security 51:50 Where to find Matt Matt Lea / Schematical: https://schematical.com Lone Wolf Programmer comics: https://schematical.com/comics Tech Debt (Steam, playtesters wanted): search "Tech Debt" on Steam (verify link) Subscribe: https://www.youtube.com/@securityunfilteredpodcast Show site: https://www.securityunfiltered.com Affiliates ➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh ➡️ OffGrid Coupon Code: JOE ➡️ Unplugged Phone: https://unplugged.com/ Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout *See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions. Tesla Referral Code: https://ts.la/joseph675128 Follow the Podcast on Social Media! Instagram: https://www.instagram.com/secunfpodcast/ Twitter: https://twitter.com/SecUnfPodcast

October 15, 2026Episode 2861 hr 4 min

Joe Grand: The Hardware Hack That Can Never Be Patched

Joe Grand, aka Kingpin, hardware hacker, L0pht alum, and the engineer behind the $2M Trezor wallet recovery, joins Joe South on Security Unfiltered. They get into how a kid with an Atari 400 ended up in the hacker underground, why the hacker mindset can't be taught, and what 23 years as an independent hacker really looks like. Joe Grand walks through learning fault injection to free a locked wallet, how he and his colleague found an attack on STM32 chips that happens before the boot ROM even runs, why a vendor that won't respin silicon leaves chips vulnerable forever, and why "they need physical access" is not a security strategy (see: the San Francisco parking meter he bought on eBay). Plus failure, burnout, full disclosure, and his new travel show exploring hacker culture around the world. Chapters 00:00 Finally reaching Joe Grand (email still wins) 03:29 Why Security Unfiltered exists 05:36 Atari 400 to the hacker underground 10:00 The hacker mindset you can't teach 13:16 Going independent for 23 years 16:30 Keep learning or get cut 21:04 Failing thousands of times with SELinux 24:16 Hacks under pressure: cameras, customers, $66M 27:55 How the $2M Trezor wallet hack started 30:46 Learning fault injection 34:37 Boot ROMs, Ghidra and silicon you can't patch 37:04 Full disclosure and fighting for the user 43:53 Chips that stay vulnerable forever 44:34 Physical access is not security: the parking meter hack 48:02 How an independent hardware hacker finds work 52:28 Saying yes, saying no, and publishing your work 56:22 Joe Grand's new hacker culture travel show 62:57 Where to find Joe Grand Joe Grand on YouTube: https://www.youtube.com/@JoeGrand Joe Grand: https://joegrand.com/ Grand Idea Studio: https://grandideastudio.com/ Subscribe: https://www.youtube.com/@securityunfilteredpodcast Show site: https://www.securityunfiltered.com Affiliates ➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh ➡️ OffGrid Coupon Code: JOE ➡️ Unplugged Phone: https://unplugged.com/ Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout *See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions. Tesla Referral Code: https://ts.la/joseph675128 Follow the Podcast on Social Media! Instagram: https://www.instagram.com/secunfpodcast/ Twitter: https://twitter.com/SecUnfPodcast

October 15, 2026Episode 28542 min

Securing AI at Scale: Why Your Logs Can't Read Agent Intent | Ashish Rajan

Ashish Rajan is back on Security Unfiltered, a few years after we last recorded together at RSA. Ashish is a CISO, host of the Cloud Security Podcast and AI Security Podcast, and author of the new Wiley book AI Security Engineering. We get into how you write a book on a field that changes every month, why agents made him rewrite his plan halfway through, and what actually matters when you have millions of agents in your environment. Ashish makes the case that logs alone can't tell you intent ("is Joe an employee?" and "can I have Joe's password?" both look like plain text), why behavior is where agent security is heading, and why no vendor is going to build the internal context layer your security team needs. We also talk open weight models vs paying for tokens, whether the government had this tech first, and why human life still comes before technology. Chapters 00:00 Ashish is back on Security Unfiltered 02:40 Joe's dissertation: zero trust for satellites and post-quantum crypto 06:36 Ashish's new book, AI Security Engineering 07:07 Ashish's path: IAM, cloud security, Cloud Security Podcast 08:26 From media company to AI advisory and the AI Security Lab 10:19 Local compute, open weight models, Grok vs Claude 13:35 Build harnesses you can swap out 14:34 How do you write a book on a topic that changes every month? 19:05 When research gets published faster than you can cite it 20:54 Why network security is back in the conversation 21:23 Agents breaking out of a sandbox 22:57 Is the government already ahead on AI? 23:54 The Manhattan Project comparison 25:13 How cyber warfare teams test exploits 26:41 Security for AI is a new field 28:16 Human life over technology: Waymo and surgical robots 29:58 Which agent security capabilities actually work? 31:40 Make the agent explain why it pulled 10 records 32:50 Behavior and intent over logs 35:12 The internal context layer no vendor will build 38:47 Logging is legacy at AI speed 40:20 Horses vs cars: who gets left behind with AI 42:17 Wrap up Ashish Rajan on LinkedIn: https://www.linkedin.com/in/ashishrajan/ AI Security Engineering (Wiley): https://www.amazon.com/dp/1394387687 TechRiot: https://techriot.io Cloud Security Podcast: https://www.cloudsecuritypodcast.tv Subscribe: https://www.youtube.com/@securityunfilteredpodcast Show site: https://www.securityunfiltered.com X: @SecUnfPodcast Affiliates ➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh ➡️ OffGrid Coupon Code: JOE ➡️ Unplugged Phone: https://unplugged.com/ Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout *See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions. Tesla Referral Code: https://ts.la/joseph675128 Follow the Podcast on Social Media! Instagram: https://www.instagram.com/secunfpodcast/ Twitter: https://twitter.com/SecUnfPodcast

October 5, 2026Episode 26549 min

Your AI Agents Will Try Every Permission You Gave Them | Nadav Cornberg Eve

SPONSORED EPISODE: This episode includes paid promotion. Security Unfiltered was paid by Eve Security for this sponsorship. Joe sits down again with Nadav Cornberg (Eve Security) for the final episode in this Eve arc. They dig into agent runtime security: understanding intent inside a session, session tainting when privileges should shrink or grow, MCP connections that let tools take action in systems the author never planned for, and why static guardrails cannot keep up once agents start chaining actions on their own. Also covered: auto-remediation in AI studios, global policy packs, and Eve's AI prevention path that pushes agentic policies into tools you already run. Part 1: https://www.youtube.com/watch?v=3oklYF1iVLs Eve Security: https://eve.security/ Nadav on LinkedIn: https://www.linkedin.com/in/nadav-cornberg/ Free AI risk assessment via Eve's contact form on eve.security. Subscribe: https://www.youtube.com/@securityunfilteredpodcast Show site: https://www.securityunfiltered.com X: @SecUnfPodcast Affiliates ➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh ➡️ OffGrid Coupon Code: JOE ➡️ Unplugged Phone: https://unplugged.com/ Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout *See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions. Tesla Referral Code: https://ts.la/joseph675128 Follow the Podcast on Social Media! Instagram: https://www.instagram.com/secunfpodcast/ Twitter: https://twitter.com/SecUnfPodcast

September 28, 2026Episode 26439 min

Why Antivirus Fails at File Scanning | OPSWAT CEO Benny Czarny on CDR

Antivirus is built to protect the device, not to judge a file. OPSWAT founder and CEO Benny Czarny explains why file scan detection lands somewhere between 5% and 40% even with AI engines, why 99.9% detection isn't good enough for a nuclear plant, and how Content Disarm and Reconstruction (CDR) protects critical infrastructure by assuming every file is malicious and rebuilding it clean. We get into: - How Benny bootstrapped OPSWAT and built a common language that let antivirus, VPN, firewall, and encryption products talk to each other - Testing every antivirus on the market with live malware, and what it showed about real-time protection vs file scanning - Why multiscanning with 30 engines still let malware through - How CDR rebuilds JPEG, PDF, and Office files to strip buffer overflow exploits before they land - Data diodes, one-way optical firewalls, and securing file flow into air-gapped networks - Offline patching, and why critical infrastructure teams need their own training (OPSWAT Academy) Benny's book, Cybersecurity Upside Down, covers the statistics and architecture behind all of this. Find Benny and OPSWAT: OPSWAT: https://www.opswat.com Benny Czarny: https://www.bennyczarny.com OPSWAT Academy: https://learn.opswatacademy.com Chapters: 00:00 Welcome Benny Czarny 02:22 How Benny got into cybersecurity 04:58 Big company vs small company lessons 08:20 First hires and scaling from generalists to experts 08:59 The problem that started OPSWAT 12:01 Testing every antivirus with live malware 13:42 Why antivirus file scanning fails, even with AI 17:36 Multiscanning file flow into critical infrastructure 19:45 Why 99.9% detection isn't good enough 20:09 Assume every file is infected: CDR explained 23:52 How CDR strips buffer overflow exploits 26:50 Dolly the sheep and distilled water analogies 28:16 From 15 file types to hundreds of formats 29:51 Data diodes and one-way optical firewalls 32:05 What CDR can't rebuild: executables and sandboxing 33:48 Walled gardens and unpatched networks 35:32 Offline patching and OPSWAT Academy 39:12 Where to find Benny and his book #cybersecurity #criticalinfrastructure #OTsecurity Affiliates ➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh ➡️ OffGrid Coupon Code: JOE ➡️ Unplugged Phone: https://unplugged.com/ Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout *See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions. Tesla Referral Code: https://ts.la/joseph675128 Follow the Podcast on Social Media! Instagram: https://www.instagram.com/secunfpodcast/ Twitter: https://twitter.com/SecUnfPodcast

September 14, 2026Episode 26353 min

100,000 OAuth Grants Later, Security Teams Still Can’t See the Risk

Joe and Danielle discuss how AI has accelerated an already messy SaaS security problem, especially as employees connect more tools, more agents, and more data without centralized oversight. The conversation focuses on why traditional block or allow controls are breaking down, and how security teams can use agentic automation to reduce risk without slowing the business 00:00 - Why AI adoption is moving faster than cloud and SaaS ever did 01:20 - The professional pressure to skill up on AI now 03:19 - How cloud security and SaaS security lagged behind adoption 06:13 - Why organizations can’t simply say no to AI tools 08:30 - The difference between cloud-era experimentation and today’s browser-based AI 10:37 - Why AI security and SaaS security are becoming the same problem 12:31 - Why traditional onboarding and vendor review processes don’t scale 14:43 - The Salesforce example that exposed hidden risk in a large enterprise 17:00 - Why most third-party risk programs only cover part of the estate 19:07 - How decentralized tech adoption bypasses security workflows 20:31 - Why OAuth grants and third-party integrations are a major attack path 22:25 - Why attackers usually go after credentials and tokens, not zero days 23:19 - How AI lowers the barrier to entry for attackers 26:55 - Why defenders need agentic capabilities too 28:16 - The scale of unmanaged OAuth grants inside enterprises 30:26 - Why no one can hire enough people to review every grant manually 31:53 - How agents can analyze risk and recommend revocation at scale 33:33 - The reality check from customer conversations about AI visibility 35:24 - How security people think when told they can’t do something 38:16 - Reactance theory and why employees work around controls 40:25 - Incentives matter more than policy slogans 41:48 - Why binary block or allow controls create shadow IT 43:30 - How Nudge Security approaches SaaS and AI as one workforce problem 46:14 - Why a workforce edge model matters more than network or identity alone 48:26 - What just-in-time policy decisions could look like in the browser 50:10 - Why policy experience is as important as policy enforcement 52:08 - Danielle on Nudge Security’s free trial and LinkedIn presence Affiliates ➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh ➡️ OffGrid Coupon Code: JOE ➡️ Unplugged Phone: https://unplugged.com/ Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout *See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions. Tesla Referral Code: https://ts.la/joseph675128 Follow the Podcast on Social Media! Instagram: https://www.instagram.com/secunfpodcast/ Twitter: https://twitter.com/SecUnfPodcast

September 8, 2026Episode 26351 min

Why AI Is Making Vulnerability Management Worse

Joe talks with Dominik Richter, co-founder of Mondu, about how AI is changing security work, vulnerability management, and the expectations placed on technical teams. The conversation digs into why so many security workflows still produce noise instead of action, and why speed without quality creates more problems than it solves. They also explore what this means for hiring, junior talent, and the skills that will matter most as AI becomes embedded in day-to-day work. 00:00 The Evolution of AI and Its Impact 06:16 Challenges in Vulnerability Management 13:00 The Role of AI in Security 20:19 The Future of Work in an AI-Driven World 21:09 Enhancing Productivity with AI 29:41 The Evolving Role of Security Professionals 40:16 Navigating the AI Landscape in Security 48:52 Adapting to a Rapidly Changing Tech Environment Affiliates ➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh ➡️ OffGrid Coupon Code: JOE ➡️ Unplugged Phone: https://unplugged.com/ Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout *See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions. Tesla Referral Code: https://ts.la/joseph675128 Follow the Podcast on Social Media! Instagram: https://www.instagram.com/secunfpodcast/ Twitter: https://twitter.com/SecUnfPodcast

August 31, 2026Episode 26353 min

The Real Innovation Wasn't Face Recognition - It Was Liveness

Joe talks with Andrew Bud, founder of iProov, about how an engineer’s career across mobile communications, payments, and identity led to one of the early liveness technologies for secure remote authentication. The conversation ranges from entrepreneurship and resilience to deepfakes, digital identity wallets, and why proving a real human is present is becoming central to cybersecurity. 00:00 The Importance of Family and Presence 03:44 Andrew Bud's Journey in Technology 09:40 The Entrepreneurial Mindset and Necessity 15:28 Conviction vs. Delusion in Entrepreneurship 23:11 Innovating Trust: The Birth of iProve 27:55 Discovering the Core Problem 29:17 The Rise of Authentication Technology 30:26 Deepfakes and the Evolution of Threats 35:19 The Challenge of Deepfake Detection 38:56 Attribution and Accountability in the Digital Age 44:51 Identity as a Defense Mechanism 49:22 Cybersecurity and National Defense Affiliates ➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh ➡️ OffGrid Coupon Code: JOE ➡️ Unplugged Phone: https://unplugged.com/ Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout *See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions. Tesla Referral Code: https://ts.la/joseph675128 Follow the Podcast on Social Media! Instagram: https://www.instagram.com/secunfpodcast/ Twitter: https://twitter.com/SecUnfPodcast

August 24, 2026Episode 2621 hr 42 min

The Real Reason Iran, Iraq, and America Keep Colliding

The conversation covers Mark's journey to the CIA, his language training, overseas assignments, and the challenges and unpredictability of his career. Mark shares insights into the preparation, language proficiency, and stress associated with overseas assignments, as well as the different roles and responsibilities within the CIA. The conversation delves into the impact of language training on CIA officers, the challenges of post-9/11 operations, the Iraq war and WMDs, and the historical context of Iran's relations with global powers. The discussion provides insights into the complexities of intelligence operations and geopolitical dynamics. The conversation delves into the history of Iran, focusing on the rise and fall of the Shahs and the country's pursuit of nuclear weapons. It explores the Shah's attempts to modernize Iran, the overthrow of Mossadegh, and the impact of the Iran-Iraq war. Additionally, it discusses the development of Iran's nuclear program and the decision-making process behind Iran's nuclear ambitions. Takeaways Language training is a crucial part of CIA preparation Overseas assignments require extensive preparation and adaptation to real-time language use Language training is crucial for CIA officers Post-9/11 operations reshaped CIA priorities and focus The Iraq war and WMDs had significant intelligence implications Iran's historical context shapes its relations with global powers The complex history of Iran's political landscape and the role of the Shahs in modernizing the country. The motivations and implications of Iran's pursuit of nuclear weapons and the challenges it presents to global security. Chapters 00:00 Introduction and Background 01:00 Journey to the Agency 03:10 Applying to the Agency 04:04 Joining the Agency 05:16 Challenges and Qualifications 06:09 Language Training and Assignments 07:26 Specializing in Farsi 08:36 Roles in the CIA 09:23 Senior Roles and Responsibilities 10:20 Career Impact and Unpredictability 11:08 Preparation for Overseas Assignments 12:05 Language Training and Deployment 13:45 Preparation for Denied Areas 14:14 Cover and Role Preparation 16:00 Language Learning and Proficiency 17:17 Adapting to Real-Time Language Use 18:14 Language Training and Stress 19:30 Preparation for Overseas Assignments 21:16 Language Training and Deployment 22:50 Language Requirements for Different Countries 24:26 Preparation for Overseas Assignments 25:43 Language Training for Denied Areas 26:58 Preparation for Overseas Assignments 28:17 Language Proficiency and Role 29:07 Adapting to Real-Time Language Use 30:39 Language Learning and Proficiency 31:57 Language Training and Adaptation 36:00 Post 9/11 Operations and Iraq 40:21 Impact of 9/11 on Operations 47:36 Iraq and WMDs 01:03:59 Iran's Historical Context 01:08:13 The History of Iran 01:18:28 Iran's Pursuit of Nuclear Weapons

August 3, 2026Episode 26149 min

The Hidden Security Risks of Agentic AI Moving Beyond Traditional Defenses

Nudge Security: https://www.nudgesecurity.com/ LinkedIn: https://www.linkedin.com/in/russell-spitler/ In this episode, Joe chats with Russell Spitler from Nudge Security about the evolving landscape of AI, SaaS, and cybersecurity. They explore how rapid technological advancements are reshaping security challenges and solutions, especially with AI's increasing capabilities and integrations. 00:00 - Russell’s cybersecurity origin story and early influences 04:24 - The industry’s rapid expansion and current security workforce trends 08:05 - How organizations struggle to manage SaaS security and AI integrations 12:49 - The shared responsibility model in SaaS and AI platforms 19:13 - How Nudge secures AI and SaaS ecosystems through API and browser insights 22:39 - Challenges with agentic AI: delegation, access, and risk mitigation 24:34 - The rise of agentic AI capabilities and security implications 27:23 - The potential for AI to democratize security and development 32:08 - Future of security automation and behavioral engagement strategies 36:58 - Building secure, scalable AI security solutions compared to container security 41:17 - The next phase: automation, policy enforcement, and behavioral redirects