The ITSPmagazine Podcast

Founded in 2015, ITSPmagazine began as a vision for a publication positioned at the critical intersection of technology, cybersecurity, and society. What started as a written publication has evolved into a comprehensive repository for all their content—podcasts, articles, event coverage, interviews, videos, panels, and everything they create. This is where Sean Martin and Marco Ciappelli talk about cybersecurity, technology, society, music, storytelling, branding, conference coverage, and whatever else catches their attention. Over a decade of conversations exploring how these worlds collide, influence each other, and shape the human experience. This is where you'll find it all.

  1. New Book! Lost in Time — Our Forgotten and Vanishing Knowledge | Forgotten Technology, Ancient Wisdom & Digital Amnesia | An Interview with Jack R. Bialik | An Analog Brain In A Digital Age With Marco Ciappelli

    10 HR AGO

    New Book! Lost in Time — Our Forgotten and Vanishing Knowledge | Forgotten Technology, Ancient Wisdom & Digital Amnesia | An Interview with Jack R. Bialik | An Analog Brain In A Digital Age With Marco Ciappelli

    New Book: Lost in Time — Our Forgotten and Vanishing Knowledge | An Interview with Jack R. Bialik | An Analog Brain In A Digital Age With Marco Ciappelli There's a particular arrogance embedded in how we talk about progress. We speak about innovation as if it moves in one direction only — forward, upward, smarter, faster. But what if the line isn't straight? What if it loops, doubles back, and occasionally vanishes entirely? That's the uncomfortable question at the center of my conversation with Jack R. Bialik. His book Lost in Time: Our Forgotten and Vanishing Knowledge doesn't read like a history lesson. It reads like a case file — evidence, example by example, that the civilization we assume is the most advanced in human history is also, in some critical ways, deeply amnesiac. Take cataract surgery. We learned it in the 1700s, right? Except we didn't. Indians were performing it in 800 BC. The ancient Egyptians and Babylonians had diagrams of the procedure dating back to 2,400 BCE. The knowledge existed, worked, and then — somewhere in the chaos of collapsing empires and burning libraries — it vanished. We didn't progress past it. We forgot it, and then reinvented it from scratch, centuries later, convinced we were doing something new. Or the Baghdad Battery: clay pots, 2,000 years old, that when filled with acid can generate 1.1 volts of electricity. We don't know what they used them for. We don't know who figured it out. We just know it worked, it existed, and then it didn't anymore. This is what Bialik calls the pattern of loss — and it's not random. It follows catastrophe: the Library of Alexandria, the systematic destruction of Mayan records, the slow erosion of oral traditions as writing systems took over. Knowledge disappears when the systems that carry it collapse. And here's where the conversation gets uncomfortably relevant: we are building those systems right now, and we are not thinking about how long they'll last. The curator at the Computer History Museum told Bialik that to preserve the data from early IBM PCs and Macintosh computers, they had to print it on paper. The floppy drives had become brittle. The formats were unreadable. The digital archive was failing — and the only solution was to go analog. A vinyl record from the 1920s still plays. A CD from the 1980s may not survive another decade. I've been thinking about this since we recorded. My brain is analog — that's not just a podcast title, it's a philosophy. I grew up in Florence, surrounded by things that had survived centuries because they were made to last: stone, fresco, manuscript. Then I jumped on the digital train like everyone else, seduced by infinite libraries on my phone, music on demand, knowledge at my fingertips. But what Bialik is pointing out is that fingertips are fragile. And so are hard drives. The deeper issue isn't storage format. It's the distinction Bialik draws between knowledge and wisdom. Knowledge is the data — the cataract surgery technique, the battery design, the pyramid engineering. Wisdom is knowing why it matters, when to use it, and what the consequences might be. We've gotten extraordinarily good at accumulating knowledge. We are considerably worse at transmitting wisdom. And wisdom, Bialik argues, doesn't live in databases. It lives in the space between people — in stories, in teaching, in the slow transmission of judgment across generations. That's why oral tradition survived when everything else failed. Not because it was more sophisticated, but because it was more human. It didn't require a device to run on. I don't know how to solve the digital longevity problem. Neither does Bialik — not yet. But I think the first step is admitting we have one. That's actually one of the quietest, most powerful arguments in the book: be humble. We don't know everything. We never did. And some of the things we've lost might be exactly what we need right now. The question isn't just what we've forgotten. It's what we're forgetting today, while we're too busy scrolling to notice. Grab Lost in Time: Our Forgotten and Vanishing Knowledge — link below — and spend some time with a perspective that goes very, very far back. Which is maybe the only way to see very, very far forward.   And if this kind of conversation is what you come here for, subscribe to the newsletter at marcociappelli.com.  More of this. Less noise. — Marco Ciappelli Co-Founder ITSPmagazine & Studio C60 | Creative Director | Branding & Marketing Advisor | Personal Branding Coach | Journalist | Writer | Podcast: An Analog Brain In A Digital Age ⚠️ Beware: Pigs May Fly | 🌎 LAX🛸FLR 🌍   ____________ About Marco Marco Ciappelli is Co-Founder & CMO of ITSPmagazine, Co-Founder & Creative Director of Studio C60, Branding & Marketing Advisor, Personal Branding Coach, Journalist, Writer, and Host of An Analog Brain In A Digital Age podcast. Born in Florence, Italy, and based in Los Angeles, he explores the intersection of technology, society, storytelling, and creativity — with an analog brain, in a digital age. 🌎 marcociappelli.com  ___________ About the Guest Jack R. Bialik is a technology expert and author with a 40-year career spanning electrical engineering, project management, F-15 fighter simulation for the U.S. Air Force, Nokia, Motorola, and the Department of Homeland Security. Lost in Time: Our Forgotten and Vanishing Knowledge is the result of years of research into the technologies, wisdom, and innovations that vanished from our collective memory — and what that means for our digital future. 🌎 jrbialik.com Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

    34 min
  2. The Autonomous SOC Is No Longer a Dream | A Brand Highlight Conversation with Subo Guha, Senior Vice President of Product Management of Stellar Cyber

    22 HR AGO

    The Autonomous SOC Is No Longer a Dream | A Brand Highlight Conversation with Subo Guha, Senior Vice President of Product Management of Stellar Cyber

    What does it take to turn the dream of an autonomous SOC into something organizations can actually deploy? Subo Guha, Senior Vice President of Product Management at Stellar Cyber, joins Sean Martin to share how the company's AI-driven security operations platform is making that vision a reality. Stellar Cyber serves SOC teams across more than 50 countries, with a primary focus on MSPs and MSSPs supporting the underserved mid-market, though marquee enterprise customers like Canon are also part of the portfolio. How can agentic AI change the way SOC teams handle alert overload? Guha describes what he calls a "digital army" of AI agents that work around the clock to automate alert triage and catch phishing attacks. The system filters 70 to 80 percent of incoming alerts, allowing analysts to focus on the 20 percent that matter most. With attackers using AI to launch faster and more frequent campaigns, Stellar Cyber takes a human-augmented approach, meaning the AI learns from analyst interactions and continuously guides the SOC team toward faster, more accurate remediation. Why does this matter for MSPs operating on thin margins? Guha explains that the autonomous SOC capability layered on top of Stellar Cyber's XDR platform allows MSSPs to serve more customers, reduce mean time to repair, and grow their tenant base without proportionally increasing staff. When MSSPs grow revenue, Stellar Cyber grows alongside them, creating a mutually beneficial model that ultimately means more organizations get protected. This is a Brand Highlight. A Brand Highlight is a ~5 minute introductory conversation designed to put a spotlight on the guest and their company. Learn more: https://www.studioc60.com/creation#highlight GUEST Subo Guha, Senior Vice President of Product Management, Stellar Cyber @LinkedIn RESOURCES Learn more about Stellar Cyber: https://stellarcyber.ai Are you interested in telling your story? ▶︎ Full Length Brand Story: https://www.studioc60.com/content-creation#full ▶︎ Brand Spotlight Story: https://www.studioc60.com/content-creation#spotlight ▶︎ Brand Highlight Story: https://www.studioc60.com/content-creation#highlight KEYWORDS Subo Guha, Stellar Cyber, Sean Martin, brand story, brand marketing, marketing podcast, brand highlight, autonomous SOC, agentic AI, security operations, XDR, NDR, MSSP, MSP, alert triage, AI-driven security, Open XDR, Gartner Magic Quadrant, phishing detection, SOC automation Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

    8 min
  3. The Operations Layer for Live Events | A Brand Highlight Conversation with Ben Ikwuagwu, CEO & Co-Founder of Soundcheck Live

    3 DAYS AGO

    The Operations Layer for Live Events | A Brand Highlight Conversation with Ben Ikwuagwu, CEO & Co-Founder of Soundcheck Live

    Ben Ikwuagwu is a vocalist, performer, and entrepreneur who has spent over 15 years navigating the live events world. That firsthand experience, combined with a degree in operations and years working in corporate America, gives him a unique vantage point on what makes the industry run and where it breaks down. Now, as CEO & Co-Founder of Soundcheck Live, he is channeling both worlds into a single platform designed to simplify how live event professionals manage their work. What does an all-in-one operations platform for live events actually do? Soundcheck Live focuses on four core pillars: booking, scheduling, payments, and coordination. Ikwuagwu explains that every event, regardless of size, comes down to these four elements. The platform provides a centralized dashboard where teams can manage gig details, client communication, and payment information without juggling spreadsheets, text threads, and scattered documents. How is Soundcheck Live building differently? From day one, the team has built the product around its users. Pilots with bands, production companies, and venues shaped the tool from the ground up. With advances in AI, the feedback loop has accelerated dramatically. Focus group insights that once took weeks to implement now translate into working features in hours, giving users the feeling that the platform is being custom-built for their specific workflows. This is a Brand Highlight. A Brand Highlight is a ~5 minute introductory conversation designed to put a spotlight on the guest and their company. Learn more: https://www.studioc60.com/creation#highlight GUEST Ben Ikwuagwu, CEO & Co-Founder of Soundcheck Live On LinkedIn: https://www.linkedin.com/in/benjaminikwuagwu/ RESOURCES Soundcheck Live (Website): https://soundchecklive.io/ Are you interested in telling your story? ▶︎ Full Length Brand Story: https://www.studioc60.com/content-creation#full ▶︎ Brand Spotlight Story: https://www.studioc60.com/content-creation#spotlight ▶︎ Brand Highlight Story: https://www.studioc60.com/content-creation#highlight KEYWORDS Ben Ikwuagwu, Soundcheck Live, Sean Martin, brand story, brand marketing, marketing podcast, brand highlight, live events, gig management, event operations, live music, booking platform, freelancer tools, event technology, live entertainment, artist management, talent agencies Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

    8 min
  4. White Knight Labs: Still 2015 — How Old Vulnerabilities and Vibe Coding Are Breaking the Future | A Brand Story Podcast John Stigerwalt Founder at White Knight Labs | Red Team Operations Leader

    4 DAYS AGO

    White Knight Labs: Still 2015 — How Old Vulnerabilities and Vibe Coding Are Breaking the Future | A Brand Story Podcast John Stigerwalt Founder at White Knight Labs | Red Team Operations Leader

    There's a particular kind of clarity you get when you talk to someone who spends their days breaking into things for a living. Not with malice — with purpose. John Steigerwald, known to most in the industry simply as "Stigs," co-founded White Knight Labs in 2016 with a mission that sounds almost disarmingly simple: build the best penetration testing team anyone has ever seen, and actually deliver results. Nearly a decade later, the company has grown to 40 people, gone international, and is busier than ever. The question worth asking is: why? The uncomfortable answer, according to Stigs, is that the fundamental problems haven't changed. At all. "Honestly, it's still 2015," he said during our most recent conversation on ITSPmagazine's Brand Story series. Not as a metaphor. As a diagnosis. The same misconfigurations, the same weak identity policies, the same unlocked back doors that red teamers were exploiting a decade ago are still wide open today. The apps built in a COVID-era frenzy — pushed out fast, tested never — are now running critical business infrastructure. And the organizations using them are only finding out when something breaks. What's changed is the surface area. Cloud, AI, Microsoft 365, vibe-coded production apps — each new layer of technology gets adopted at speed, and each one arrives carrying the same original sin: no one turned on the basics. Stigs used Microsoft 365 as a pointed example. Millions of businesses are running on it with DMARC turned off, default configurations untouched, Copilot layered on top, and not a single CIS Benchmark policy applied. "Every client is vulnerable," he said. "Not just 10% of clients. Every client." That's a striking statement. It's also, if you've been paying attention to breach headlines, not a surprising one. The AI angle adds a new and almost darkly comedic wrinkle. Vibe coding — the practice of using AI tools like Cursor or Claude to generate production-ready code at speed — has given entry-level developers intermediate-level output. Which sounds great, until you realize that the AI models many of them leaned on were trained on outdated, sometimes vulnerable data. Stigs described visiting multiple clients with nearly identical security weaknesses, all tracing back to the same ChatGPT-generated setup instructions. "You and your neighbor did the same thing," he told one client. That's not just a funny anecdote. It's a warning about what happens when an entire industry bootstraps its infrastructure from the same flawed source. And yet, Stigs isn't anti-AI. He uses it every day. He just sees it with the clarity of someone who also finds the holes it leaves behind. His prediction for the near future: a massive wave of secure code review requests, as companies start reckoning with the vibe-coded backlog they've been quietly accumulating. AppSec is about to have a very good year. Looking forward, White Knight Labs is watching the growing intersection of private sector expertise and government infrastructure testing with particular interest. Critical infrastructure in America, long overdue for rigorous physical and embedded testing, is starting to receive that attention. Stigs and his team are already in the room. What makes White Knight Labs different isn't just technical skill — it's the ability to communicate what they find in language that actually lands. In an industry full of reports that gather dust, that matters. The best penetration test in the world is useless if no one acts on it. The door is open. It's been open for years. The question is who you call to finally lock it. To learn more about White Knight Labs, visit their website or reach out directly. Listen to the full conversation on ITSPmagazine. GUEST John Stigerwalt Founder at White Knight Labs | Red Team Operations Leader https://www.linkedin.com/in/john-stigerwalt-90a9b4110/ RESOURCES White Knight Labs:  https://whiteknightlabs.com _____________________________________________________________ Are you interested in telling your story? ▶︎ Full Length Brand Story: https://www.studioc60.com/content-creation#full ▶︎ Brand Spotlight Story: https://www.studioc60.com/content-creation#spotlight ▶︎ Brand Highlight Story: https://www.studioc60.com/content-creation#highlight Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

    37 min
  5. The New Identity Risk AI Agents Bring to the Enterprise | A Brand Highlight Conversation with Ido Shlomo, Co-Founder & CTO of Token Security

    4 DAYS AGO

    The New Identity Risk AI Agents Bring to the Enterprise | A Brand Highlight Conversation with Ido Shlomo, Co-Founder & CTO of Token Security

    What happens when AI agents inherit access to enterprise systems but nobody governs their identities? Ido Shlomo, Co-Founder and CTO of Token Security, joins the conversation to unpack a rapidly growing challenge that many organizations face but few have addressed. As businesses accelerate AI adoption, agents are being deployed to fetch data from CRMs, process emails, and execute actions across platforms. The problem is that these agents often operate with persistent access, no clear ownership, and little visibility into what they can reach. How should security teams approach AI agent identity governance? Shlomo explains that the first step is discovery. Most companies do not know what their AI agent inventory looks like, and without that baseline, effective governance is impossible. The good news, he notes, is that agents do not suffer from politics. They do exactly what they are told and operate within the boundaries they are given. That predictability makes the challenge more manageable if the right tooling is in place. What makes an effective access policy for AI agents? Rather than relying on prompt filtering or output controls that add latency and friction, Shlomo advocates for intent-based permission models that scope each agent to access only what it needs, when it needs it. He frames the prioritization process as a matrix of access and autonomy, where the agents with the highest levels of both deserve immediate attention. For business leaders, the visibility that comes from this approach also reveals waste and inefficiency, highlighting departments and services that are not delivering on their intended value. To learn more about how to identify, govern, and secure AI agent identities, connect with the Token Security team and follow Ido Shlomo for practical guidance. This is a Brand Highlight. A Brand Highlight is a ~5 minute introductory conversation designed to put a spotlight on the guest and their company. Learn more: https://www.studioc60.com/creation#highlight GUEST Ido Shlomo, Co-Founder & CTO of Token Security On LinkedIn: https://il.linkedin.com/in/ido--shlomo RESOURCES Token Security (Website): https://www.token.security/ Are you interested in telling your story? ▶︎ Full Length Brand Story: https://www.studioc60.com/content-creation#full ▶︎ Brand Spotlight Story: https://www.studioc60.com/content-creation#spotlight ▶︎ Brand Highlight Story: https://www.studioc60.com/content-creation#highlight KEYWORDS Ido Shlomo, Token Security, Sean Martin, brand story, brand marketing, marketing podcast, brand highlight, AI agent identity, non-human identity, identity governance, AI agent security, identity risk, least privilege, AI agent access, machine identity, NHI security, AI agent inventory, intent-based access Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

    7 min
  6. Agade: The AI-Powered Wearable Robots That Protect Workers, Not Replace Them | A Brand Highlight Conversation with Lorenzo Aquilante, Co-Founder and AGADE

    14 FEB

    Agade: The AI-Powered Wearable Robots That Protect Workers, Not Replace Them | A Brand Highlight Conversation with Lorenzo Aquilante, Co-Founder and AGADE

    Agade: The AI-Powered Wearable Robots That Protect Workers, Not Replace Them  AI Meets Human Craftsmanship There's something poetic about a technology born to help people with muscular dystrophy finding its second life on factory floors and logistics warehouses. That's the story of Agade, an Italian deeptech startup that began as a research project at Politecnico di Milano and evolved into something far more ambitious: a mission to preserve human craftsmanship in an age of automation. I sat down with Lorenzo Aquilante, CEO and co-founder of Agade, to talk about their journey from healthcare innovation to industrial exoskeletons—and what it was like showcasing their latest product at CES 2026. The origin story matters here. Back in 2017, researchers at Politecnico di Milano started developing exoskeletons for people affected by muscular dystrophy. They created something different—a semi-active model powered by AI that recognizes when a user is lifting and responds accordingly. It wasn't just about motors and sensors. It was about intelligence. Then companies came knocking. Manufacturing firms, logistics operations, industries where human workers still matter because their skills, experience, and judgment can't be replaced by machines. They saw potential. Why not use this technology to protect the people doing the heavy lifting—literally? Agade was founded in 2020 with a clear mission: preserve craftsmanship against the physical toll of material handling. Not replace humans. Protect them. The company now has two products. The first, launched in 2024, focuses on shoulder assistance. The second—the one they brought to CES 2026—targets the lower back, which makes sense when you consider that back pain is practically an occupational hazard for anyone moving materials all day. What makes Agade's approach different is that semi-active AI system. The exoskeleton knows when you're lifting. It responds. It's not just a passive brace or a fully motorized suit that takes over. It's somewhere in between—smart enough to help, light enough to wear all day. Lorenzo emphasized something that resonated with me: the importance of feedback. From day one, Agade has been obsessed with real-world testing. Not lab conditions. Actual workers doing actual jobs. Because the buyer isn't the user—companies purchase these for their employees—and that creates a unique dynamic. You need both sides to believe in the technology. The CES experience brought that home. There's always the initial wow factor when someone sees a wearable robot with motors and sensors. But the real work happens after the demo, when users tell you what needs to improve. That's where the collaboration lives. And here's what struck me most about this conversation: Agade isn't trying to remove humans from the equation. They're trying to keep humans in it longer, healthier, and more capable. In a world racing toward full automation, there's something refreshing about a company betting on human skill—and building technology to protect it. The products are available globally. You can reach Agade through their website at agadexoskeletons.com, find them on LinkedIn and other social channels, and even arrange trials before committing to a purchase. For those of us watching the intersection of AI, robotics, and human labor, Agade represents a different path. Not humans versus machines. Humans with machines. Tools that amplify rather than replace. That's a story worth telling. Marco Ciappelli interviews Lorenzo Aquilante, CEO & Co-Founder of Agade, for ITSPmagazine's Brand Highlight series following CES 2026. >>> Marcociappelli.com GUEST Lorenzo Aquilante, CEO and co-founder of Agade https://www.linkedin.com/in/lorenzo-aquilante-108573b0/ RESOURCES AGADE: https://agade-exoskeletons.com Are you interested in telling your story? ▶︎ Full Length Brand Story: https://www.studioc60.com/content-creation#full ▶︎ Brand Spotlight Story: https://www.studioc60.com/content-creation#spotlight ▶︎ Brand Highlight Story: https://www.studioc60.com/content-creation#highlight KEYWORDS Agade, exoskeleton, CES 2026, wearable robotics, AI, future of work, industrial exoskeleton, made in Italy, workplace safety, deeptech, robotics. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

    7 min
  7. KEVology: How Exploit Scores and Timelines Shape Real Security Decisions | A Brand Highlight Conversation with Tod Beardsley, Vice President of Security Research of runZero

    13 FEB

    KEVology: How Exploit Scores and Timelines Shape Real Security Decisions | A Brand Highlight Conversation with Tod Beardsley, Vice President of Security Research of runZero

    The CISA Known Exploited Vulnerabilities (KEV) catalog is one of the most referenced resources in vulnerability management, but how well do security teams actually understand what it tells them? In this Brand Highlight, Tod Beardsley, Vice President of Security Research at runZero and former CISA section chief who helped manage the KEV on a daily basis, breaks down what the catalog is designed to do and, just as importantly, what it is not. What is the KEV catalog and who is it really for? The KEV is mandated by Binding Operational Directive 22-01 (BOD 22-01), which tasks CISA with identifying vulnerabilities that are known to be exploited and have an available fix. Its primary audience is federal civilian executive branch agencies, but because the catalog is public, organizations everywhere use it as a prioritization signal. Beardsley notes that inclusion on the KEV requires a CVE ID, evidence of active exploitation, a patch or mitigation, and relevance to federal interests, meaning zero-day vulnerabilities and end-of-life systems without CVEs never appear. How should organizations think about KEV entries that are not equally dangerous? Beardsley explains that only about a third of KEV-listed vulnerabilities represent straight-shot remote code execution with no user interaction and no authentication required. The rest span a wide spectrum of severity. EPSS data reveals an inverse bell curve: many KEV entries have extremely low probabilities of exploitation in the next 30 days, while others cluster at the high end with commodity exploits widely available. This means treating every KEV entry as equally critical leads to wasted effort and alert fatigue. That gap between the catalog and real-world decision-making is exactly what KEVology addresses. The research, produced by Beardsley at runZero, enriches KEV data with CVSS metrics, EPSS scores, exploit tooling indicators, and ATT&CK mappings to help security teams filter and prioritize vulnerabilities based on what actually matters to their environment. Rather than prescribing a single priority list, KEVology treats the KEV as data to be analyzed, not doctrine to be followed blindly. To make this analysis accessible and interactive, runZero built KEV Collider, a free, daily-updated web application at runzero.com/kev-collider. The tool lets defenders sort, filter, and layer multiple risk signals across the entire KEV catalog. Because every filter combination is encoded in URL parameters, teams can bookmark and share custom views with colleagues instantly. Beardsley describes KEV Collider as an evergreen companion to the research, updating automatically as new vulnerabilities are added to the catalog each week. This is a Brand Highlight. A Brand Highlight is a ~5 minute introductory conversation designed to put a spotlight on the guest and their company. Learn more: https://www.studioc60.com/creation#highlight GUEST Tod Beardsley, Vice President of Security Research at runZero On LinkedIn: https://www.linkedin.com/in/todb/ RESOURCES Learn more about runZero: https://www.runzero.com KEVology research report: https://www.runzero.com/resources/kevology/ KEV Collider: https://www.runzero.com/kev-collider/ Are you interested in telling your story? ▶︎ Full Length Brand Story: https://www.studioc60.com/content-creation#full ▶︎ Brand Spotlight Story: https://www.studioc60.com/content-creation#spotlight ▶︎ Brand Highlight Story: https://www.studioc60.com/content-creation#highlight KEYWORDS Tod Beardsley, runZero, Sean Martin, brand story, brand marketing, marketing podcast, brand highlight, KEVology, KEV Collider, CISA KEV, vulnerability management, exploit scoring, EPSS, CVSS, vulnerability prioritization, exposure management, BOD 22-01, known exploited vulnerabilities, cybersecurity risk, patch management Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

    8 min
  8. Semantic Chaining: A New Image-Based Jailbreak Targeting Multimodal AI | A Brand Highlight Conversation with Alessandro Pignati, AI Security Researcher of NeuralTrust

    13 FEB

    Semantic Chaining: A New Image-Based Jailbreak Targeting Multimodal AI | A Brand Highlight Conversation with Alessandro Pignati, AI Security Researcher of NeuralTrust

    What happens when AI safety filters fail to catch harmful content hidden inside images? Alessandro Pignati, AI Security Researcher at NeuralTrust, joins Sean Martin to reveal a newly discovered vulnerability that affects some of the most widely used image-generation models on the market today. The technique, called semantic chaining, is an image-based jailbreak attack discovered by the NeuralTrust research team, and it raises important questions about how enterprises secure their multimodal AI deployments. How does semantic chaining work? Pignati explains that the attack uses a single prompt composed of several parts. It begins with a benign scenario, such as a historical or educational context. A second instruction asks the model to make an innocent modification, like changing the color of a background. The final, critical step introduces a malicious directive, instructing the model to embed harmful content directly into the generated image. Because image-generation models apply fewer safety filters than their text-based counterparts, the harmful instructions are rendered inside the image without triggering the usual safeguards. The NeuralTrust research team tested semantic chaining against prominent models including Gemini Nano Pro, Grok 4, and Seedream 4.5 by ByteDance, finding the attack effective across all of them. For enterprises, the implications extend well beyond consumer use cases. Pignati notes that if an AI agent or chatbot has access to a knowledge base containing sensitive information or personal data, a carefully structured semantic chaining prompt can force the model to generate that data directly into an image, bypassing text-based safety mechanisms entirely. Organizations looking to learn more about semantic chaining and the broader landscape of AI agent security can visit the NeuralTrust blog, where the research team publishes detailed breakdowns of their findings. NeuralTrust also offers a newsletter with regular updates on agent security research and newly discovered vulnerabilities. This is a Brand Highlight. A Brand Highlight is a ~5 minute introductory conversation designed to put a spotlight on the guest and their company. Learn more: https://www.studioc60.com/creation#highlight GUEST Alessandro Pignati, AI Security Researcher, NeuralTrust On LinkedIn: https://www.linkedin.com/in/alessandro-pignati/ RESOURCES Learn more about NeuralTrust: https://neuraltrust.ai/ Are you interested in telling your story? ▶︎ Full Length Brand Story: https://www.studioc60.com/content-creation#full ▶︎ Brand Spotlight Story: https://www.studioc60.com/content-creation#spotlight ▶︎ Brand Highlight Story: https://www.studioc60.com/content-creation#highlight KEYWORDS Alessandro Pignati, NeuralTrust, Sean Martin, brand story, brand marketing, marketing podcast, brand highlight, semantic chaining, image jailbreak, AI security, agentic AI, multimodal AI, LLM safety, AI red teaming, prompt injection, AI agent security, image-based attacks, enterprise AI security Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

    7 min

About

Founded in 2015, ITSPmagazine began as a vision for a publication positioned at the critical intersection of technology, cybersecurity, and society. What started as a written publication has evolved into a comprehensive repository for all their content—podcasts, articles, event coverage, interviews, videos, panels, and everything they create. This is where Sean Martin and Marco Ciappelli talk about cybersecurity, technology, society, music, storytelling, branding, conference coverage, and whatever else catches their attention. Over a decade of conversations exploring how these worlds collide, influence each other, and shape the human experience. This is where you'll find it all.

More From ITSPmagazine Podcasts