Full Segment Notes
AI agents can independently discover and install new tools, but the emerging ecosystem of AI Skills and MCP servers lacks many of the trust and security controls applied to traditional software. Michael Leland will discuss Island research uncovering thousands of malicious repositories, widespread security flaws across MCP servers, and a new attack technique called “AgentBaiting,” in which attackers manipulate agents into finding and recommending malware to users. The conversation will explore how enterprises can govern AI capabilities without slowing adoption.Segment Resources:https://www.island.io/blog/agentbaiting-how-800-fake-ai-skills-and-mcp-servers-delivered-malwareFor more information about Island's research, please visit: https://securityweekly.com/islandbh
Guest
Michael Leland brings over 30 years of data networking, operations, and cybersecurity domain expertise. He formerly served as Head of Technical Marketing and Chief Cybersecurity Evangelist at SentinelOne where he was responsible for messaging and strategic development of their XDR product roadmap as well as the identity security portfolio. Prior to SentinelOne, he held the title of Chief Technical Strategist for McAfee. Michael was the co-founder and CTO of NitroSecurity – later acquired by McAfee – where he was responsible for developing and implementing their overall SIEM technology vision and roadmap and has held senior technical management positions at Cabletron and Avaya.
