WTF Friday (aka Jargon Alert): AI Agent

By Patrick Diamond, CFP®

Artificial Intelligence is hard to escape these days. I keep hearing the term AI Agent and wasn’t sure what that even meant really… so here’s a basic explanation.

A large language model (LLM) is the “brain” behind tools (also known as chatbots) like ChatGPT and Google Gemini. Its primary job is to understand text and generate human-like responses. These LLMs are excellent at generating and editing text, but, for example, don’t have access to personal data like your Outlook or Google calendar. It has been trained on massive amounts of written information so it can answer questions, summarize documents, write emails, explain concepts, and carry on conversations. However, on its own, an LLM is passive. It responds when you ask something, but it does not take independent action (like, for example, checking your calendar). Think of it as a very advanced autocomplete system that predicts the most useful next words based on your input.

An AI agent, on the other hand, is a system that uses an LLM as its reasoning engine but adds the ability to take action. An agent can follow multi-step instructions, make decisions, use external tools (such as calendars, web browsers, weather apps, or spreadsheets), and complete tasks on your behalf. For example, instead of just explaining how to book a flight, an AI agent could compare prices, select an option based on your preferences, and prepare the booking details. In simple terms: the LLM provides intelligence and language understanding, while the AI agent adds autonomy and execution. All AI Agents have to be able to reason and act. What people in the AI world call the “ReAct Framework.” The combination turns AI from a “smart assistant that answers questions” into a “digital worker that can get things done.”

*This blog post is for informational and educational purposes only. The information contained in this blog post is not legal, tax, or investment advice.