BROADCASTS.com

On this week's episode of The Microsoft Threat Intelligence Podcast, Sherrod DeGrippo is joined by Mark Russinovich.\xa0 Mark Russinovich, CTO and Technical Fellow of Microsoft Azure, joins the show to talk about his journey from developing on-prem tools like Sysinternals to working in the cloud with Azure. Sherrod and Mark discuss the evolution of cybersecurity, the role of AI in threat intelligence, and the challenge of jailbreaking AI models. Mark shares his experiences with testing AI models for vulnerabilities, including his discovery of the "Crescendo" and "Masterkey" methods to bypass safety protocols. They also touch on the issue of poisoned training data and its impact on AI reliability, while highlighting the importance of staying ahead in cybersecurity.\xa0\n\n\nIn this episode you\u2019ll learn:\xa0\xa0\xa0\xa0\xa0\xa0\n\nThe shift from desktop computing to cloud-based systems and its implications\xa0\n\nPotential consequences of AI models having overridable safety instructions\xa0\n\nHow AI training data can manipulate the outcomes generated by AI models\xa0\n\n\n\nSome questions we ask:\xa0\xa0\xa0\xa0\xa0\n\nWill AI owners be able to stop data poisoning, or will it become more common?\xa0\n\nCan you share challenges and vulnerabilities in maintaining the security of AI systems?\xa0\n\nWhat sparked your interest in AI jailbreaks, and what trends are you seeing?\xa0\n\n\n\nResources:\xa0\xa0\nView Mark Russinovich on LinkedIn\xa0\xa0\nView Sherrod DeGrippo on LinkedIn \xa0\n\xa0\nAI jailbreaks: What they are and how they can be mitigated?\nhttps://www.microsoft.com/en-us/security/blog/2024/06/04/ai-jailbreaks-what-they-are-and-how-they-can-be-mitigated/\xa0\n\nInside AI Security with Mark Russinovich | BRK227\xa0\nhttps://www.youtube.com/watch?v=f0MDjS9-dNw\xa0\nHow Microsoft discovers and mitigates evolving attacks against AI guardrails.\nhttps://www.microsoft.com/en-us/security/blog/2024/04/11/how-microsoft-discovers-and-mitigates-evolving-attacks-against-ai-guardrails/\xa0\nGoogle AI said to put glue on pizza.\nhttps://www.businessinsider.com/google-ai-glue-pizza-i-tried-it-2024-5\xa0\n\xa0\n\n\ufeffRelated Microsoft Podcasts:\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\xa0\n\n\nAfternoon Cyber Tea with Ann Johnson\xa0\n\n\nThe BlueHat Podcast\xa0\n\n\nUncovering Hidden Risks\xa0\xa0\xa0\xa0\xa0\n\n\xa0\nDiscover and follow other Microsoft podcasts at microsoft.com/podcasts\xa0\xa0\nGet the latest threat intelligence insights and guidance at Microsoft Security Insider\xa0\n\xa0\nThe Microsoft Threat Intelligence Podcast is produced by Microsoft and distributed as part of N2K media network.