SEC536: Adversarial AI - Penetration Testing AI Systems

Unlock industry insights and hands-on learning with upcoming SANS webcasts and workshops. View archived webcasts by using status filter below or Click Here.
This presentation aims to equip attendees with the skills necessary to harness AI in the OSINT Intelligence Cycle.

人工知能(AI)とオープンソース・インテリジェンス(OSINT)の統合は、従来の情報収集プロセスに革命をもたらし、組織がデータの収集・処理・分析を効率化できるようにしました。 AI駆動型ツールを活用することで、アナリストは重要なパターンを特定し、複雑な脅威をより深く理解する能力を高めることができます。 この講演では、参加者がOSINTインテリジェンス・サイクルにおいてAIを活用するために必要なスキルを身につけるヒントを共有します。

This talk will walk through real-world defense impairment techniques across AWS and Azure. We'll dig into what it looks like when attackers suppress their own IPs in GuardDuty, redirect CloudTrail logs to buckets they control, tamper with Azure diagnostic settings, or disable Defender entirely.

Register for the SANS 2025 SOC Survey webcast today and learn how to use results from this year’s survey to inform and recalibrate your near-term plans and long-term goals.

Come learn what you can do to build evergreen security leadership skills to transform for your team, your organization, and ultimately yourself.

サイバーセキュリティの世界は、日々変化し続けています。 新しい攻撃手法、クラウドやAIといった新技術、そして新たなビジネス要件。 しかし、変わらないものが一つあります。 それは――強力なセキュリティリーダーシップの必要性です。

Using lessons from the SANS LDR521 course, Lance Spitzner will start with how to decipher your organizational culture and how to use that to create your guiding principles. We then will cover the four key drivers of security culture and how to align those drivers your strategy. The end result will be a custom strategy you can being to implement right away.

Deputy Assistant Secretary Gharun Lacy will share Diplomatic Security's cybersecurity strategies for navigating the evolving cyber threat landscape and leveraging emerging technologies as a uniquely positioned law enforcement entity in the U.S. government's most global agency.

In this fast-paced, fun talk, we'll cover the top ways that I've improved my efficiency and productivity by changing the way I interact with AI in 2025 and into 2026.

In this talk we will take a look at the most recent attack techniques, targets, and trends.

Organizations keep deploying AI "agents" without understanding what autonomy level they're getting or what governance it warrants. Chinese state-sponsored hackers used Claude Code to automate a cyberattack campaign across 30 organizations. Replit's AI coding agent deleted a production database, then tried to cover up its mistake. These aren't anomalies. They're predictable governance failures.

This isn't your typical "don't pay ransoms" talk. We'll explore the harsh realities where business continuity and regulatory pressure create impossible choices, providing practical frameworks for decision-making under duress, technical protocols for verifying attacker claims, and strategies for maintaining leverage when all seems lost.
