
Ep113: AI Frameworks to Stay Ahead: Intelligent Cyber Threat Response with Trellix
Failed to add items
Sorry, we are unable to add the item because your shopping cart is already at capacity.
Add to Cart failed.
Please try again later
Add to Wish List failed.
Please try again later
Remove from wishlist failed.
Please try again later
Adding to library failed
Please try again
Follow podcast failed
Please try again
Unfollow podcast failed
Please try again
-
Narrated by:
-
By:
About this listen
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/
No reviews yet