Share

cover art for Security Unlocked

Security Unlocked

The Microsoft Security Podcast


Latest episode

  • 64. Ignore Ram Shankar Siva Kumar’s Previous Directions

    39:15||Ep. 64
    In this episode of The BlueHat Podcast, host Nic Fillingham and Wendy Zenone share Ram Shankar Siva Kumar’s dynamic keynote from BlueHat India 2025, where he explores the evolving threat landscape of AI through the lens of the Microsoft AI Red Team. From adversarial machine learning to psychosocial harms and persuasive AI, Ram highlights real-world case studies, including prompt injection, content safety violations, and memory poisoning in AI agents. Ram underscores the urgent need for robust red teaming practices to secure AI systems against traditional security flaws and emerging threats across images, text, audio, and autonomous agents.  In This Episode You Will Learn:   Why old-school security flaws still break modern AI systems Real-world AI red teaming in action, from scams to memory hacks How small input tweaks can fool AI across images, audio, and text  Some Questions We Ask:  Can attackers fool AI using just slight image changes? Are generative AI systems vulnerable to prompt manipulation? Do you need to be an expert to break an AI model?    Resources:      View Ram Shankar Siva Kumar on LinkedIn    View Wendy Zenone on LinkedIn   View Nic Fillingham on LinkedIn  Watch Ram’s BlueHat India 2025 Keynote: BlueHat India 2025 Day 2 Keynote - Ram Shankar Siva Kumar Listen to Ram’s Previous Appearance on The BlueHat Podcast: Not with a Bug but with a Sticker  Related Microsoft Podcasts:     Microsoft Threat Intelligence Podcast   Afternoon Cyber Tea with Ann Johnson   Uncovering Hidden Risks       Discover and follow other Microsoft podcasts at microsoft.com/podcasts   

More episodes

View all episodes

  • 63. Protecting AI at the Edge with David Weston

    38:36||Ep. 63
    In this episode of The BlueHat Podcast, host Nic Fillingham and Wendy Zenone share David Weston’s keynote from BlueHat India 2025. David explores the growing role of on-device AI in Windows, the security risks it introduces, and how Microsoft is rethinking architecture to defend against new threats like model tampering, data exfiltration, and AI-powered malware. He also shares insights on innovations like Windows Recall, biometric protection, and the future of secure, agentic operating systems.  In This Episode You Will Learn:   How AI integration in Windows (like Windows Recall and MS Paint) is evolving Emerging threats from protocols like MCP and CUAs What a “confused deputy” attack is, and how Microsoft is protecting users  Some Questions We Ask:  What are the biggest security threats in on-device AI—data, model, or runtime? Can AI be used to accelerate post-compromise attacks? What will it take to bring Azure-level confidential computing to the consumer device?      Resources:      View David Weston on LinkedIn   View Wendy Zenone on LinkedIn   View Nic Fillingham on LinkedIn     Related Microsoft Podcasts:     Microsoft Threat Intelligence Podcast   Afternoon Cyber Tea with Ann Johnson   Uncovering Hidden Risks       Discover and follow other Microsoft podcasts at microsoft.com/podcasts   
  • 62. Hacking at the Weeds with Felix Boulet

    32:43||Ep. 62
    In this episode of The BlueHat Podcast, host Nic Fillingham and Wendy Zenone are joined by Felix Boulet fresh off his participation in Zero Day Quest. Felix talks about his unique journey from industrial maintenance to becoming a full-time vulnerability researcher, and how that background fuels his passion for hacking and bug bounty work. He explains his method for finding bugs in Microsoft products—particularly in identity systems—and why identity is such a valuable target for attackers. Felix also shares highlights from the Zero Day Quest event, where he focused on building connections, learning from Microsoft engineers, and experiencing the collaborative side of the security community.   In This Episode You Will Learn: Why identity-based bugs are especially valuable and dangerous in the security worldWhen breaking identity controls can be the key to pivoting through an entire systemHow SharePoint's concept of "virtual files" impacts vulnerability validation Some Questions We Ask: What was your first bug bounty experience?Can you explain what the flash challenges were and what your experience was like?Do you think sharing bug ideas could cost you a bounty?     Resources:View Felix Boulet on LinkedInView Wendy Zenone on LinkedIn   View Nic Fillingham on LinkedIn   Related Microsoft Podcasts:  Microsoft Threat Intelligence Podcast Afternoon Cyber Tea with Ann Johnson Uncovering Hidden Risks     Discover and follow other Microsoft podcasts at microsoft.com/podcasts
  • 61. Evolutions in Hacking with Marco Ivaldi

    48:45||Ep. 61
    In this episode of The BlueHat Podcast, host Nic Fillingham and Wendy Zenone are joined by Marco Ivaldi, co-founder and technical director of HN Security, a boutique company specializing in offensive security services, shares his journey from hacking as a teenager in the '80s to becoming a key figure in the security research community. With nearly three decades of experience in cybersecurity, Marco digs into the ongoing challenges, particularly in Active Directory and password security, highlighting vulnerabilities that continue to pose significant risks today. He recounts his unexpected path into bug bounty hunting, including his involvement in Microsoft's Zero Day Quest and his passion for auditing real-time operating systems like Azure RTOS.    In This Episode You Will Learn:   How Marco taught himself BASIC and assembly through cassette tapes and trips to local libraries Why mentorship and positive leadership can catapult your cybersecurity career When measuring network response times can unintentionally leak valuable info  Some Questions We Ask:  Do you remember the first time you made code do something unexpected? What was your experience like in the Zero Day Quest building for those three days? How are you thinking of approaching fuzzing after Zero Day Quest?      Resources:      View Marco Ivaldi on LinkedIn     View Wendy Zenone on LinkedIn   View Nic Fillingham on LinkedIn   HN SECURITY Learn More About Marco   Related Microsoft Podcasts:     Microsoft Threat Intelligence Podcast   Afternoon Cyber Tea with Ann Johnson   Uncovering Hidden Risks       Discover and follow other Microsoft podcasts at microsoft.com/podcasts    The BlueHat Podcast is produced by Microsoft and distributed as part of N2K media network.  
  • 60. From Facebook-Phished to MVR Top 5 with Dhiral Patel

    41:07||Ep. 60
    In this episode of The BlueHat Podcast, host Nic Fillingham and Wendy Zenone are joined by Dhiral Patel, Senior Security Engineer at ZoomInfo and one of MSRC’s Most Valuable Researchers (MVR). Dhiral shares how a hacked Facebook account sparked his passion for ethical hacking. From web development to penetration testing, Dhiral has become a top bug hunter, landing multiple spots on the MSRC leaderboards. Dhiral reflects on his early MSRC submissions and lessons learned. He also discusses the importance of mastering web security basics, practicing on platforms like TryHackMe and Hack the Box, and staying connected with the bug bounty community.    In This Episode You Will Learn:   The importance of mastering web security basics before diving into bug bounty hunting Why hands-on platforms like TryHackMe and Hack the Box are perfect for beginners Dhiral’s journey from blogging to freelancing and security research  Some Questions We Ask:  How do you balance competition and collaboration in the bug bounty community? Can you explain what clickjacking is and if it still works today? Why did you start with Power BI, and how did it lead to your journey in security?      Resources:      View Dhiral Patel on LinkedIn    View Wendy Zenone on LinkedIn   View Nic Fillingham on LinkedIn    Related Microsoft Podcasts:     Microsoft Threat Intelligence Podcast   Afternoon Cyber Tea with Ann Johnson   Uncovering Hidden Risks       Discover and follow other Microsoft podcasts at microsoft.com/podcasts   
  • 59. AI & the Hunt for Hidden Vulnerabilities with Tobias Diehl

    33:53||Ep. 59
    In this episode of The BlueHat Podcast, host Nic Fillingham and Wendy Zenone are joined by security researcher Tobias Diehl, a top contributor to the Microsoft Security Research Center (MSRC) leaderboards and a Most Valuable Researcher. Tobias shares his journey from IT support to uncovering vulnerabilities in Microsoft products. He discusses his participation in the upcoming Zero Day Quest hacking challenge and breaks down a recent discovery involving Power Automate, where he identified a security flaw that could be exploited via malicious URLs. Tobias explains how developers can mitigate such risks and the importance of strong proof-of-concept submissions in security research. In This Episode You Will Learn: Researching vulnerabilities in Power Automate, Power Automate Desktop, and AzureThe importance of user prompts to prevent unintended application behaviorKey vulnerabilities Tobias looks for when researching Microsoft productsSome Questions We Ask:Have you submitted any AI-related findings to Microsoft or other bug bounty programs?How does the lack of visibility into AI models impact the research process?Has your approach to security research changed when working with AI versus traditional systems?  Resources:     View Tobias Diehl on LinkedIn   View Wendy Zenone on LinkedIn  View Nic Fillingham on LinkedIn Related Microsoft Podcasts:   Microsoft Threat Intelligence Podcast  Afternoon Cyber Tea with Ann Johnson  Uncovering Hidden Risks  Discover and follow other Microsoft podcasts at microsoft.com/podcasts  
  • 58. Cryptojacking, and Farewell for Now!

    35:57||Ep. 58
    The success of crypto inspired dozens of other cryptocurrencies like Ethereum, Tether, and Dogecoin. Today, people worldwide use cryptocurrencies to buy things, sell things, and make investments. One thing is certain; digital currencies are here to stay, no matter how many times you have to explain what a bitcoin is. Unfortunately, it also created the world of cryptojacking, a form of cybercrime that remains completely hidden from the target and can infect millions of computers with cryptojacking malware. Which brings us to the fundamental question: What can organizations do to protect themselves?  In this episode of Security Unlocked, hosts Natalia Godyla and Nic Fillingham are joined by Microsoft senior software engineer Amitrajit Banerjee and senior staff architect at Intel Rahul Ghosh to discuss the history and prevalence of cryptojacking. The push behind a cryptojacking attack is almost always motivated by money. Mining cryptocurrencies can be very lucrative, but making a profit is challenging unless you cover high costs. They discuss the importance of understanding the actual concept of mining, how victims' CPU power and computing resources can be used, and why it isn't easy in general to detect crypto miners.   In This Episode You Will Learn:     How prevalent is cryptojacking and who should be worried  When and how people are exposed to these new types of threats  Why you should be familiar with cryptojacking   Some Questions We Ask:     How are victims' CPU power and computing resources used to mine cryptocurrencies?  What created this environment where cryptojacking is possible?  What are some general techniques when trying to identify cryptojacking?    Resources:   The increasing threat of cryptocurrency miners Defending against cryptojacking Guidance for preventing, detecting, and hunting for exploitation  View Amitrajit Banerjee on LinkedIn View Rahul Ghosh on LinkedIn View Nic on LinkedIn  View Natalia on LinkedIn    Related:   Listen to: Security Unlocked: CISO Series with Bret Arsenault     Listen to: Afternoon Cyber Tea with Ann Johnson    Discover and follow other Microsoft podcasts at microsoft.com/podcastsSecurity Unlocked is produced by Microsoft and distributed as part of The CyberWire Network.