How many AI agents does an enterprise actually need? Box CTO Ben Kus breaks down how to think about the agentic ecosystem and why the answer is not one. Most enterprise teams now have access to more AI agents than they know what to do with. Frontier agents like Claude, ChatGPT, and Gemini on one side. Platform agents built into the tools that hold your data on the other. The question is not which one to use. It is how to use both together. In this short explainer, Ben Kus, CTO of Box, breaks down the distinction between generalist and specialist agents and explains why the most sophisticated enterprise AI architectures do not rely on a single agent to do everything. Frontier agents are generalists. They can use MCP servers, write code, and handle a wide range of knowledge and developer work. Platform agents like Box AI are specialists. They are built to find, analyze, and act on unstructured data inside a specific platform in ways a generalist agent cannot replicate on its own. The insight: you do not have to choose. A frontier agent can reach into Box headlessly to retrieve content. Or it can go further and ask Box AI to do the specialized work, create a document, generate a report, analyze a folder, and hand the result back. The generalist orchestrates. The specialist executes. That is what a mature agentic architecture looks like. Not one agent doing everything. Multiple agents cooperating, each doing what it does best. This is part of Box's AI Explainer Series: short, practical videos that help enterprise teams understand how to get the most out of AI in their workflows. FAQs: Q: How many AI agents does an enterprise actually need? A: According to Ben Kus, the better question is not how many agents you need but how they work together. Most enterprises have access to both generalist frontier agents and specialist platform agents. The most effective architectures use both, with generalist agents orchestrating and specialist agents executing. Q: What is the difference between a generalist agent and a specialist agent? A: Generalist agents like Claude, ChatGPT, and Gemini are designed to handle a wide range of tasks including using MCP servers, writing code, and performing knowledge work. Specialist agents like Box AI are built to excel at specific tasks within a specific platform, such as finding, analyzing, and acting on unstructured data in Box. Q: Can frontier agents and platform agents work together? A: Yes. A frontier agent can reach into Box headlessly to retrieve content, or it can delegate specialized work to Box AI directly, asking it to create a document, generate a report, or analyze a folder, and then use that output as part of a larger workflow. Q: What is a headless agent integration? A: A headless integration means an AI agent can access and interact with a platform like Box directly through its APIs or MCP server, without requiring a human interface. This allows agents to retrieve, analyze, and act on enterprise content programmatically as part of automated workflows. Q: How does Box fit into an agentic ecosystem? A: Box serves as the governed content layer that any AI agent can connect to. Whether a frontier agent is retrieving files headlessly through the Box MCP Server or delegating specialized tasks to Box AI, your existing permissions, governance, and security controls are enforced automatically throughout.

 96600      4