AI agent infrastructure may be a bigger business than agents

The AI agent market may not be only about building agents directly. Agents need many supporting tools to work in real products, including marketplaces, usage analytics, monitoring, security, payments, and tools for agent-to-agent communication. They also need infrastructure for deploying and managing many agents.

The idea is similar to a gold rush, where the companies selling tools can sometimes do better than the people searching for gold. The central question is whether the biggest opportunity is in agent products themselves or in the systems that help agents run safely and reliably.

Key points

Quick term guide

marketplace
An online place where many sellers and buyers can meet and trade.
monitoring
Watching a system to see if it is working well or having problems.
agent-to-agent communication
A way for separate AI agents to exchange information or coordinate work.
infrastructure
The technical systems that keep a website or app running.
model call
One request sent to an AI model to get an answer.
visibility
How easily people can discover and notice a product.
deployment
The process of putting software changes into a running system.
monitoring tool
Software that checks whether an app, website, or server is working normally.
Read original