From Prompting to Autonomy: The Evolution of Adversarial AI

Aurora Blum and Rohit Nambiar from Google's Threat Intelligence Group join host Luke McNamara to discuss their research into agentic pentesting frameworks such as ARTEX. They detail the progression this year from prompt-based usage of LLMs to autonomous frameworks that require little operator interaction, and how this is changing the speed and scale of attacks. https://cloud.google.com/blog/topics/threat-intelligence/artex-and-friends-enter-agentic-offensive-frameworks

Authors' of GTIG's AI Threat Tracker report, Aurora Blum and Kelli Vanderlee, join host Luke McNamara to discuss the latest observations from the Q3 version of this report. Aurora and Kelli note the shift to more agentic use cases, as actors evolve being individual prompting, the rise of LLMjacking and supply chain attacks that have impacted internal AI infrastructure and resources for enterprises, and more. https://cloud.google.com/blog/topics/threat-intelligence/from-prompting-to-autonomy-the-evolution-of-adversarial-ai

In this episode of Mandiant’s Defender’s Advantage Podcast, host Luke McNamara sits down with Ben Read, Head of Strategic Threat Intelligence at Wiz, to explore the rapidly shifting landscape of software supply chain compromises. While historic, nation-state operations like SolarWinds focused on compromising closed-source software, modern adversaries have expanded their playbook to target widely used open-source ecosystems, repositories, and automated CI/CD pipelines. Ben discusses how differing tactics in these campaigns play out in the current threat landscape. For more on Wiz's research: https://www.wiz.io/blog/tag/research 

Host Luke McNamara sits down with Muhammad Muneer, Technical Manager on Mandiant's Incident Response team, to unpack the stark realities of enterprise AI adoption. They explore why securing AI starts with resolving legacy security debt (specifically around IAM, supply chain, and secrets management) and dissect the critical differences between "governance for AI" and "governance of AI." Muhammad details real-world frontline findings—from developers bypassing API gateways to the financial costs of leaked LLM API keys—and outlines the emerging threat of adversaries leveraging LLM-enabled command-line tools for living-off-the-land attacks.

Host Luke McNamara is joined by Jake Nicastro, who leads the AI function for the Frontline Intelligence Operations team within the Google Threat Intelligence Group (GTIG). Jake details how his team is shifting from simple prompt engineering to more advanced agentic workflows, focusing on a model of "human-machine teaming." He shares practical use cases for AI in CTI—including automated script decoding, hunting query creation, and streamlining repetitive metadata-labeling tasks. Jake also discusses the concept of "judge agents" for quality control, the implementation of structured analytic techniques, and how real-time AI assistants can accelerate on-boarding and domain transitions for frontline threat analysts.

Host Luke McNamara is joined by Charley Snyder, Head of Disruption Operations at Google Threat Intelligence Group, to delve into how Google is crafting a more coordinate approach to disrupting adversary cyber operations. Charley describes how this disruption focus is not hacking back, how it builds on existing work across Google security teams, and some of the recent wins such as the IPIDEA and GRIDTIDE takedowns. 

Host Luke McNamara is joined by Chris Linklater, Practice Leader at Mandiant, to discuss the 2026 edition of Mandiant's M-Trends Report. Chris dives into the latest trends observed in breached throughout 2025 and into this year, noting some of the key aspects organizations should focus on in applying these insights into today's threat landscape. https://cloud.google.com/security/resources/m-trends

In this episode of the Defenders Advantage Podcast, host Luke McNamara sits down with Google Threat Intelligence experts Jose Nazario and Brandon Wood. They dive into the rollout of new dark web and underground monitoring capabilities, explaining how AI is fundamentally changing the way defenders track adversaries.https://cloud.google.com/blog/products/identity-security/bringing-dark-web-intelligence-into-the-ai-era\

Host Luke McNamara is joined by Eugene Liderman, Senior Director in Android's Security and Privacy Group, to discuss the evolving world of mobile-targeting scams. Eugene details some of the unique aspects to mobile scams, regional variations in tactics by scammers, and the steps Android has taken to combat this problem. 

Sarah Yoder (Manager, Mandiant Consulting) and Ashley Pearson (Senior Analyst, Advanced Practices on Google Threat Intelligence Group) join host Luke McNamara to discuss UNC5221 and their operations involving BRICKSTORM backdoor. This highly sophisticated, suspected China-nexus cyber-espionage threat group is known for aggressively targeting internet-facing network appliances (like VPNs and firewalls) to establish long-term, stealthy access for espionage.Read our blog post for more: https://cloud.google.com/blog/topics/threat-intelligence/brickstorm-espionage-campaign