Ep071: Protecting Your Enterprise in the Era of Generative AI with Sanjay Kalra of Zscaler

Sanjay Kalra of Zscaler and Randy DeFauw of Amazon Web Services explore the hidden dangers of generative AI security—from invisible text manipulation and deep fakes to data poisoning and dark AI models—while offering practical strategies for protecting your enterprise in this era of generative AI.Topics Include:AI security threats grouped into data, malicious use, trust/safetyData security critical for SaaS-based AI servicesModel training data vulnerable to poisoning and manipulationGenAI lacks traditional data deletion capabilitiesAccess controls difficult once data becomes model embeddingsPrompt injection attacks becoming widespread, with libraries available onlineDeepfake scams increasing in sophistication and frequencyAI enhancing phishing attacks with better written contentDark AI models emerging specifically for malicious purposesModel hallucinations being exploited for security attacksAI accelerating analysis of stolen dataShadow AI usage by employees poses security risksExisting vendor AI integration creating unexpected security challengesFine-grained access controls essential for AI applicationsPII protection critical in both inputs and outputsComprehensive prompt and response logging necessaryInvisible text manipulation emerging in resumes and RFPsModel fine-tuning can compromise built-in security guardrailsMulti-language inputs create new security considerationsCompetition-sensitive content requires careful AI managementAI firewalls needed for input/output monitoringRegular security testing required for AI modelsAI compliance standards emerging globallyMulti-modal AI creating new security challengesBrowser isolation helping control AI application usageParticipants:Sanjay Kalra – Product Management at ZscalerRandy DeFauw – Senior Principal Solutions Architect, Amazon Web ServicesSee how Amazon Web Services gives you the freedom to migrate, innovate, and scale your software company at https://aws.amazon.com/isv/

There are no comments yet.

Please log in to write the first comment.