{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"<iframe src=\"https://embed.acast.com/$/64ac54ede55ebb0011cb41b1/6840cf46c8835d385c4c1e7c?\" frameBorder=\"0\" width=\"700\" height=\"250\"></iframe>","title":"Ignore Ram Shankar Siva Kumar’s Previous Directions ","description":"<p>In this episode of The&nbsp;BlueHat&nbsp;Podcast, host&nbsp;<a href=\"https://www.linkedin.com/in/nicfill/\" rel=\"noopener noreferrer\" target=\"_blank\">Nic Fillingham</a>&nbsp;and&nbsp;<a href=\"https://www.linkedin.com/in/wendyzenone/\" rel=\"noopener noreferrer\" target=\"_blank\">Wendy Zenone</a>&nbsp;share&nbsp;<a href=\"https://www.linkedin.com/in/rssk/?utm_source=chatgpt.com\" rel=\"noopener noreferrer\" target=\"_blank\">Ram Shankar Siva Kumar’s</a>&nbsp;dynamic keynote from&nbsp;BlueHat&nbsp;India&nbsp;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.&nbsp;Ram&nbsp;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.&nbsp;</p><p>&nbsp;</p><p><br></p><p><strong>In This Episode You Will Learn</strong>:&nbsp;&nbsp;</p><p>&nbsp;</p><ul><li>Why old-school security flaws still break modern AI systems&nbsp;</li><li>Real-world AI red teaming in action, from&nbsp;scams&nbsp;to memory hacks&nbsp;</li><li>How small input tweaks can fool AI across images, audio, and text&nbsp;</li></ul><p>&nbsp;</p><p><strong>Some Questions We Ask:</strong>&nbsp;</p><p>&nbsp;</p><ul><li>Can attackers fool AI using just slight image changes?&nbsp;</li><li>Are generative AI systems vulnerable to prompt manipulation?&nbsp;</li><li>Do you need to be an expert to break an AI model?&nbsp;</li></ul><p>&nbsp;&nbsp;&nbsp;</p><p><strong>Resources:</strong>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</p><p><a href=\"https://www.linkedin.com/in/rssk/?utm_source=chatgpt.com\" rel=\"noopener noreferrer\" target=\"_blank\">View Ram Shankar Siva Kumar on LinkedIn</a>&nbsp;&nbsp;&nbsp;&nbsp;</p><p><a href=\"https://www.linkedin.com/in/wendyzenone/\" rel=\"noopener noreferrer\" target=\"_blank\">View Wendy Zenone on LinkedIn</a>&nbsp;&nbsp;&nbsp;</p><p><a href=\"https://www.linkedin.com/in/nicfill/\" rel=\"noopener noreferrer\" target=\"_blank\">View Nic Fillingham on LinkedIn</a>&nbsp;&nbsp;</p><p><br></p><p>Watch&nbsp;Ram’s&nbsp;BlueHat&nbsp;India 2025 Keynote:&nbsp;<a href=\"https://www.youtube.com/watch?v=SalQD4Kio0U\" rel=\"noopener noreferrer\" target=\"_blank\">BlueHat India 2025 Day 2 Keynote - Ram Shankar Siva Kumar</a>&nbsp;</p><p>Listen&nbsp;to Ram’s Previous Appearance on The&nbsp;BlueHat&nbsp;Podcast:&nbsp;<a href=\"https://thecyberwire.com/podcasts/the-bluehat-podcast/10/notes\" rel=\"noopener noreferrer\" target=\"_blank\">Not with a Bug but with a Sticker</a>&nbsp;</p><p>&nbsp;</p><p><br></p><p><strong>Related Microsoft Podcasts:</strong>&nbsp;&nbsp;&nbsp;</p><p>&nbsp;&nbsp;</p><ul><li><a href=\"https://msthreatintelpodcast.com/\" rel=\"noopener noreferrer\" target=\"_blank\">Microsoft Threat Intelligence Podcast</a>&nbsp;&nbsp;&nbsp;</li><li><a href=\"https://afternooncybertea.com/\" rel=\"noopener noreferrer\" target=\"_blank\">Afternoon Cyber Tea with Ann Johnson</a>&nbsp;&nbsp;&nbsp;</li><li><a href=\"https://uncoveringhiddenrisks.com/\" rel=\"noopener noreferrer\" target=\"_blank\">Uncovering Hidden Risks</a>&nbsp;&nbsp;&nbsp;</li></ul><p>&nbsp;&nbsp;</p><p>&nbsp;&nbsp;</p><p>Discover and follow other Microsoft podcasts at<a href=\"https://news.microsoft.com/podcasts/\" rel=\"noopener noreferrer\" target=\"_blank\">&nbsp;microsoft.com/podcasts</a>&nbsp;&nbsp;&nbsp;</p>","author_name":"Microsoft"}