Ep113: AI Frameworks to Stay Ahead: Intelligent Cyber Threat Response with Trellix Podcast Por  arte de portada

Ep113: AI Frameworks to Stay Ahead: Intelligent Cyber Threat Response with Trellix

Ep113: AI Frameworks to Stay Ahead: Intelligent Cyber Threat Response with Trellix

Escúchala gratis

Ver detalles del espectáculo

Acerca de esta escucha

Wilson Patton, Solutions Architect for Trellix, demonstrates how their four-pillar Gen-AI framework transforms incident alerts into actionable intelligence.

Topics Include:

  • Wilson Patton: Trellix Solutions Architect, 20 years government experience
  • Witnessed evolution from basic firewalls to zero trust architectures
  • Trellix combines McAfee and FireEye heritage and capabilities
  • AI integration isn't new - machine learning embedded for years
  • Partnership with AWS Bedrock accelerates Gen-AI development capabilities
  • 2014: Developed Impossible Travel Analytic for anomaly detection
  • 2016: Launched Guided Investigations framework for SOC analysts
  • 2023: Introduced AI Guided Investigations with contextual understanding
  • 64% of public sector exploring AI adoption actively
  • Only 21% have requisite data ready for training
  • Gen-AI won't magically clean up messy, siloed data
  • 74% of executives doubt AI information accuracy currently
  • Monday morning alert queue: 76 high, 318 medium alerts
  • Adversaries steal credentials 90 days before major incidents
  • Critical breadcrumbs hidden in low-priority informational alerts
  • 1000+ data-driven investigative questions developed over eight years
  • Skilled analysts take too long reading all answers
  • Automate analysis, distill thousands down to ten critical alerts
  • Four foundational pillars for effective, trustworthy Gen-AI implementation
  • Cybersecurity expertise essential - Gen-AI is just a tool
  • Frameworks ensure reliability and consistent prompting for production
  • Multiple LLM models tested through AWS Bedrock platform
  • Quality diverse datasets required for accurate question answering
  • Good prompts combine evidence, context, and comprehensive information
  • Testing shows order of magnitude price differences between models
  • Nova Micro provides cost-effective results for many scenarios
  • Prompt engineering superior to fine-tuning for avoiding bias
  • Agentic AI performs multi-step investigations with live data
  • Strategic model choice based on specific requirements and costs
  • Transparent audit trails mandatory for government compliance requirements


Participants:

  • Wilson Patton – Solutions Architect, Trellix


Further Links:

  • Website: https://www.trellix.com
  • Trellix in the AWS Marketplace


See how Amazon Web Services gives you the freedom to migrate, innovate, and scale your software company at https://aws.amazon.com/isv/

Todavía no hay opiniones