A company is building an AI orchestration layer that coordinates multiple AI agents to complete complex tasks, including browsing the web, reading emails, and executing code. What is the MOST significant new risk category introduced by this multi-agent architecture?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Here's the deal — b is correct because multi-agent systems introduce a new attack surface where a single compromised, hallucinating, or prompt-injected agent can trigger a cascade of actions across other agents, potentially escalating privileges and taking irreversible real-world actions at scale. An agent with web browsing capability that receives a malicious instruction can trigger code execution in another agent, which then sends emails via a third agent.
Full explanation below image
Full Explanation
B is correct because multi-agent systems introduce a new attack surface where a single compromised, hallucinating, or prompt-injected agent can trigger a cascade of actions across other agents, potentially escalating privileges and taking irreversible real-world actions at scale. An agent with web browsing capability that receives a malicious instruction can trigger code execution in another agent, which then sends emails via a third agent. This cascading amplification of errors is unique to agentic architectures. Computational cost (A), IP risk (C), and context limitations (D) are real concerns but not the defining new risk category of multi-agent systems.