The WestCap CISO Summit brings security leaders to New York City for three days of talks, workshops, and networking focused on the ideas shaping modern defense. Hosted by WestCap, a growth equity firm with cybersecurity among its core investment areas, the summit gives CISOs and practitioners a chance to compare notes with peers working through the same challenges. Bishop Fox is glad to be part of this year's conversation.

On Wednesday, October 21, Thomas Wilson, Senior Red Team Operator at Bishop Fox, joins Adam Malone, Global Head, Acute Events for Cyber Risk; Avi Sambria, Vice President of North America at Sygnia; and Stephan Goldberg, Co-Founder & Chief Product Officer at ACT Security for "Preparing for an AI Attack." Thomas brings an attacker's perspective on agentic AI: what adversaries look for when organizations hand autonomy to software, and why those systems leave defenders so little signal to work with. His view is shaped by hands-on AI/LLM security assessments and red teaming.

If you're responsible for securing AI adoption across your organization, catch the panel and find Thomas afterward to talk shop about how attackers are approaching agentic systems today.

For more details, visit: WestCap CISO Summit

"Preparing for an AI Attack"

Speakers: Thomas Wilson, Senior Red Team Operator, Bishop Fox

Date/Time: Wednesday, October 21 | 3:00 PM ET

Abstract: AI agents now read inboxes, query databases, call APIs, and move data on behalf of the people who deploy them. Every one of those actions is something the agent is supposed to do, which is exactly what makes the agentic attack surface so hard to defend. When a manipulated or compromised agent abuses its access, the activity looks nearly identical to the legitimate work around it. Detection programs were built to find the needle in the haystack. With agents, malicious behavior hides in a needle stack, buried among thousands of other needles that look just like it.

In this panel, Thomas hosts security leaders from Kroll, Sygnia, and ACT Security will examine what preparing for AI-driven attacks looks like in practice. Thomas will unpack how attackers turn trusted agent behavior to their advantage, why conventional detection struggles to separate legitimate agent activity from malicious use, and where security leaders can build visibility and guardrails before an incident forces the issue. Attendees will leave with a sharper picture of where agentic risk concentrates and the right questions to bring back to their teams.