Securing the Invisible Thread: Threat Modeling and Defending MCP in Agentic AI Systems
This AI Learning Series session by Vandana Verma explores how MCP connects LLMs to tools, APIs, data, and systems, quietly reshaping trust boundaries and expanding the attack surface beyond what traditional AppSec models were built for.
As agentic AI adoption accelerates, the session focuses on how MCP can turn well-intentioned agents into effective attack orchestrators through:
- context poisoning
- token and secret leakage
- privilege escalation via scope creep
- tool misuse
- shadow MCP servers
The goal of the session is not to slow AI innovation, but to secure how power flows across agentic systems.
The event timing is listed as 31 January 2026 at 8:00 PM IST.
Reference
https://lnkd.in/gUfwHx2d
https://www.linkedin.com/posts/vandana-verma_aisecurity-agenticai-mcp-activity-7423333828800364545-SyjO