Episodes

  • Your AI Policy Is Not an AI Security Program | Jason Loomis | Ep 118
    Nov 16 2026

    Send us Fan Mail

    AI is moving faster than most organizations can govern it.

    Employees are already using multiple models. AI agents are gaining access to systems and data. Products are being built with AI embedded into core workflows. Yet many organizations still treat an AI policy as if it were an AI security program.

    It isn’t.

    In this episode of Musings from the Cyber Trench, I sit down with cybersecurity executive Jason Loomis to discuss what AI security looks like from the CISO seat—and why the hardest problems may have less to do with the technology than with leadership, accountability, access, and execution.

    We get into the practical realities of securing AI and agentic systems, including:

    • Why executives often have a very different picture of AI usage than what is actually happening inside the organization
    • Why publishing an AI policy is only the first step toward meaningful governance
    • How familiar security principles like MFA, least privilege, data inventory, and classification still apply to AI
    • Why agentic AI creates new challenges around permissions, changing intent, task-scoped credentials, and time-bound access
    • Which guardrails should be non-negotiable before AI can influence sensitive data, decisions, or customer outcomes
    • Why startups and fast-moving companies often defer security—and the downstream cost of trying to bolt it on later
    • Why executives remain accountable when AI makes or influences a bad decision
    • How GRC teams can keep pace as AI frameworks, controls, products, and risks evolve faster than traditional governance cycles
    • Whether AI can eventually monitor and govern other AI systems—and who watches the system doing the monitoring
    • Why AI-generated software still requires human judgment, product thinking, security review, and an understanding of how people actually work

    One theme keeps surfacing throughout the conversation: AI does not eliminate accountability.

    If an AI agent has excessive permissions, leadership owns that risk. If an AI-enabled product creates an unacceptable outcome, leadership cannot simply say, “AI did it.” And if an organization waits until AI is already deeply embedded in production before bringing security into the conversation, remediation becomes significantly harder.

    Jason’s advice to leaders is straightforward: start now.

    You do not need to solve every AI security problem on day one. But you should know where AI is being used, understand what it can access, identify your highest risks, involve the people closest to the technology, and begin putting guardrails in place before an incident forces the conversation.

    ABOUT JASON LOOMIS

    Jason Loomis is an experienced cybersecurity executive and CISO whose background spans SaaS, healthcare, banking, finance, e-commerce, mergers and acquisitions, IPO and pre-IPO environments.

    His work sits at the intersection of cybersecurity, organizational decision-making, risk, and leadership. In this conversation, Jason also discusses how his experiences outside cybersecurity—including disaster relief work and adventure sports—have influenced the way he thinks about leadership, resilience, and risk.

    ZERO TRUST READINESS ASSESSMENT

    Responsible for ICAM, Zero Trust, or identity security within a federal agency, prime contractor, or large regulated enterprise?

    If you are trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment:

    https://zephon.tech/zt

    Questions or guest suggestions?

    defend@zephon.tech

    Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?

    If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt

    Questions or guest ideas? Email defend@zephon.tech

    Show More Show Less
    54 mins
  • Why AI Projects Fail: The Technology Usually Isn’t the Problem | Hao Dinh | Episode 117
    Nov 2 2026

    Send us Fan Mail

    Most AI projects do not fail because the model is not powerful enough. They fail because organizations bolt AI onto broken processes and expect the technology to fix everything.

    Successful AI adoption still comes back to the fundamentals: people, process, technology, ownership, trust, good data, and a clear business outcome.

    In this episode of Musings from the Cyber Trench, I sit down with Hao Dinh, an AI and digital transformation practitioner who has spent more than a decade working with AI and now focuses heavily on helping non-technical professionals understand how to use it.

    We discuss practical AI adoption at both the individual and enterprise level. Hao breaks down why AI initiatives stall, how organizations can integrate AI into real workflows, and why the goal should not be deploying more AI—it should be helping people do useful work better.

    Hao shares practical insights on:

    • Why AI adoption becomes easier when people can apply it to familiar tasks rather than learning technology in the abstract
    • Why enterprise AI projects stall when teams prioritize model performance but ignore people, workflows, trust, and adoption
    • How the BOLTED framework identifies six recurring failure patterns in AI initiatives
    • How the HARMONY framework connects human oversight, workflow design, responsible AI, measurable impact, ownership, data, and organizational context
    • Why AI ROI should be measured through business outcomes rather than technical accuracy alone
    • How non-technical professionals can use AI to automate repetitive work and spend more time on higher-value activities
    • Why teams should start by asking employees what parts of their jobs are painful before deciding where AI belongs
    • How vibe coding and citizen programming may allow employees to solve narrowly defined workflow problems without immediately buying another software platform
    • Why heavy AI use can improve productivity while potentially weakening the boredom and creative struggle that often produce original thinking

    The central takeaway is simple: AI works best as an enabler, not as the strategy itself. Start with the human problem. Understand the workflow. Fix what is broken. Then determine where AI can make that process faster, easier, or more valuable.

    And as individuals, we need the same balance. AI can help us overcome the blank page and explore possibilities, but we still need room to think, struggle, experiment, and occasionally be bored.

    ABOUT HAO DINH

    Hao Dinh is an AI and digital transformation practitioner with more than a decade of experience working with AI and machine learning. He is also an Aspen Institute fellow and works through Grow by Design to help non-technical professionals build practical AI skills. His background includes digital transformation, professional skateboarding, and a strong focus on making emerging technology accessible and useful to everyday professionals.

    ZERO TRUST READINESS ASSESSMENT

    Responsible for ICAM, Zero Trust, or identity security within a federal agency, prime contractor, or large regulated enterprise?

    If you are trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment:

    https://zephon.tech/zt

    Questions or guest suggestions?

    defend@zephon.tech

    Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?

    If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt

    Questions or guest ideas? Email defend@zephon.tech

    Show More Show Less
    57 mins
  • Cybersecurity Skills Are Changing Faster Than Our Hiring Models | Anil Saxena | Episode 116
    Oct 19 2026

    Send us Fan Mail

    AI is changing cybersecurity work. But the bigger challenge for leaders may be whether their talent strategy is changing fast enough to keep up.

    The organizations that succeed will not simply hire more experienced people. They will need to identify potential earlier, develop talent internally, create room for continuous learning, and rethink what entry-level cybersecurity work should look like in an AI-enabled environment.

    In this episode of Musings from the Cyber Trench, I sit down with Anil Saxena, a workforce development and platform-driven learning leader at ServiceNow, to discuss what is broken in the way organizations hire, train, and develop cybersecurity talent.

    We explore how AI is changing cybersecurity roles, why credentials and years of experience are incomplete measures of potential, how simulation-based learning could improve technical hiring, and why organizations need to treat cybersecurity talent as a long-term system rather than a series of positions to fill.

    Anil shares practical insights on:

    • Why organizations need to develop cybersecurity talent internally instead of relying primarily on experienced external hires
    • Why analytical thinking, pattern recognition, curiosity, and stakeholder management become more valuable as AI automates routine work
    • How leaders can evaluate cybersecurity potential beyond certifications and years of experience
    • Why simulation-based exercises may provide a better way to assess how candidates solve real problems
    • How AI can eliminate low-value administrative work without eliminating the need for meaningful entry-level roles
    • Why cybersecurity training fails when organizations use one-size-fits-all programs without follow-up or hands-on application
    • How organizations can build a cybersecurity talent pipeline around future business needs instead of reacting to vacancies
    • Why mid-career cybersecurity professionals should learn to work with AI rather than compete against it
    • How strong technical teams depend on leadership, shared working norms, and a common objective

    The central takeaway is that cybersecurity workforce development cannot remain reactive. Organizations need to think several years ahead, identify people with the right underlying capabilities, give them meaningful opportunities to learn, and deliberately build the bench before the talent is urgently needed.

    ABOUT ANIL SAXENA

    Anil Saxena works in workforce development and platform-driven learning at ServiceNow, including work with ServiceNow University. His background spans engineering, organizational design, enablement, and the development of technical teams. His work focuses on how organizations build skills, develop people, and create learning systems that can scale as technology and workforce requirements change.

    ZERO TRUST READINESS ASSESSMENT

    Responsible for ICAM, Zero Trust, or identity security within a federal agency, prime contractor, or large regulated enterprise?

    If you are trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment:

    https://zephon.tech/zt

    Questions or guest suggestions?

    defend@zephon.tech

    Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?

    If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt

    Questions or guest ideas? Email defend@zephon.tech

    Show More Show Less
    59 mins
  • Have We Shifted Cybersecurity Too Far Left? | Rick Clark | Episode 115
    Sep 30 2026

    Send us Fan Mail

    Moving security left was supposed to reduce risk. Instead, many organizations have made developers responsible for hundreds of controls they cannot realistically understand, manage, or consistently enforce.

    In this episode of Musings from the Cyber Trench, I sit down with Rick Clark, a cloud, open-source, Linux, and platform engineering leader with three decades of experience, to discuss whether cybersecurity has shifted too far left.

    We examine how organizations can build security controls directly into their platforms, reduce control sprawl, respond faster to vulnerabilities, and measure software delivery based on business and security outcomes, not deployment speed alone.

    Rick shares practical insights on:

    • Embedding security controls into CI/CD platforms so they are applied automatically
    • Keeping control ownership with information security while reducing the burden on developers
    • Restoring architectural oversight without unnecessarily slowing delivery
    • Reviewing legacy controls instead of carrying them unchanged from on-premises environments into the cloud
    • Using AI to enforce security controls rather than simply producing code faster
    • Measuring software delivery against business value, compliance, and risk
    • Recovering from zero-day vulnerabilities through standardized containers and centralized platform control
    • Using mutual TLS and end-to-end identity tracking to secure platform communications
    • Protecting sensitive data, private keys, and certificates from immature AI agents

    The central takeaway is straightforward: developers should write secure code, but they cannot carry the organization’s entire cybersecurity, operations, compliance, and cost-management burden. Security teams must define the controls, and platform teams must make those controls consistent, automatic, and easier to use.

    ABOUT RICK CLARK

    Rick Clark has 30 years of experience across cloud computing, open source, Linux, platform engineering, and security. He led the development of Ubuntu Server, led engineering for the Rackspace Cloud and helped found OpenStack, worked in the CTO’s office at Cisco Systems, and helped build the Reliance Jio cloud infrastructure in India.

    ZERO TRUST READINESS ASSESSMENT

    Responsible for ICAM, Zero Trust, or identity security within a federal agency, prime contractor, or large regulated enterprise?

    If you are trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment:

    https://zephon.tech/zt

    Questions or guest suggestions?

    defend@zephon.tech

    Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?

    If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt

    Questions or guest ideas? Email defend@zephon.tech

    Show More Show Less
    1 hr and 3 mins
  • Prompt Injection Is a Data Access Problem | Sundar Krishnamurthy | EP 114
    Sep 14 2026

    Send us Fan Mail

    You cannot secure AI by filtering prompts alone. If an AI agent can retrieve sensitive data or execute actions, prompt injection becomes an identity, authorization, and architecture problem.

    In this episode of Musings from the Cyber Trench, I sit down with Sundar Krishnamurthy, Senior Security Architect at Expedia Group, to discuss how organizations can secure AI in real enterprise environments.

    We examine why prompt-level guardrails are not enough, how non-human identities should remain connected to the people they represent, and why security controls must be enforced close to the data and action being requested.

    Sundar shares practical insights on:

    • Turning security assumptions into automated tests
    • Using LLMs to support code reviews and threat modeling
    • How indirect prompts can bypass basic guardrails
    • Enforcing authorization at the point of data retrieval
    • Maintaining human and non-human identity context
    • Isolating vector stores to reduce cross-tenant exposure
    • Avoiding security shortcuts during MVP development
    • Protecting private data when using AI models

    The central message is simple: AI security cannot depend on trusting the model or recognizing every malicious prompt. Identity, authorization, data isolation, logging, and defense in depth must be built into the underlying architecture.

    ABOUT SUNDAR KRISHNAMURTHY

    Sundar Krishnamurthy is a Senior Security Architect at Expedia Group. He conducts architecture reviews across applications, cloud, networks, infrastructure, and AI and machine learning security. His background includes work with AWS, SAP Concur, and Microsoft.

    Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?

    If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt

    Questions or guest ideas? Email defend@zephon.tech

    Show More Show Less
    59 mins
  • Compliance, GRC, cybersecurity maturity, audit readiness, AI, CMMC, and continuous security
    May 31 2026

    Send us Fan Mail

    Too many organizations still treat compliance as a one-time audit exercise: get the certification, satisfy the customer, and move on.

    In this episode of Musings from the Cyber Trench, I sit down with Sarah Lynn, a seasoned IT, cybersecurity, GRC, advisory, and audit preparation leader, to discuss why that mindset breaks down fast.

    We talk about what happens when compliance is treated as “paperwork,” where programs usually fail first, and why people, process, and technology all have to work together for compliance to become part of daily operations.

    Sarah also shares practical insights on:

    • Why undocumented processes are a major red flag
    • How leaders can move from checklist compliance to security maturity
    • Where organizations underinvest and overspend in compliance programs
    • Why buying a tool before understanding the process usually backfires
    • AI’s role in compliance, automation, meeting notes, artifact collection, and risk
    • Why CMMC, SOC 2, ISO, FedRAMP, and other frameworks require continuous effort
    • How trusted advisors and peer groups can help leaders avoid reinventing the wheel

    The core message: compliance is not something you “get through.” Done right, it becomes a habit, a management discipline, and a foundation for stronger security.

    Guest: Sarah Lynn brings 25+ years across IT, cybersecurity, GRC, audit readiness, risk, continuity, and technology operations, helping SaaS/IaaS-driven organizations turn compliance into practical, business-aligned security.

    Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?

    If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt

    Questions or guest ideas? Email defend@zephon.tech

    Show More Show Less
    55 mins
  • Energy Is the Upstream Cyber Risk in Power Infrastructure | Bethun Bhowmik | EP 112
    Apr 22 2026

    Send us Fan Mail

    Energy is the upstream of everything

    In this episode we explore how energy infrastructure has become one of the most critical and vulnerable systems in the modern world

    From power grid attacks to systemic risks this conversation reveals why cyber threats to energy impact entire nations

    Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?

    If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt

    Questions or guest ideas? Email defend@zephon.tech

    Show More Show Less
    1 hr
  • The Leadership Mistake That Breaks Security Systems | Morgan Reed | EP 111
    Apr 8 2026

    Send us Fan Mail

    This episode explores the leadership and design challenges behind modern cybersecurity failures.

    Morgan Reed, CTO of Transbridge, shares how traditional approaches to security focused on controls, compliance, and restriction often ignore the most critical variable: human behavior.

    The discussion reframes cybersecurity as a design and leadership problem, where usability, context, and adaptability determine effectiveness.

    You’ll learn: - Why leadership decisions shape security outcomes - How excessive controls create friction and risk - The gap between security policy and real world behavior - Why human centered design is critical in cybersecurity - How AI can support adaptive, context aware systems - What leaders must change to build resilient security environments

    This episode is ideal for executives, CISOs, and technology leaders focused on improving security, reducing risk, and building systems that actually work in practice.

    Responsible for ICAM, Zero Trust, or identity security in a federal agency, prime, or large regulated enterprise?

    If you’re trying to move from strategy to execution, start with Zephon’s Zero Trust Readiness Assessment: zephon.tech/zt

    Questions or guest ideas? Email defend@zephon.tech

    Show More Show Less
    57 mins