AI trading comes with a lot of jargon. This AI trading glossary explains 30 terms in plain English, so you can read about trading agents without getting lost.
AI trading glossary: the terms
Agentic AI — AI systems that can perceive their environment, make decisions, and take actions on their own, rather than just answering questions.
AI Trading Bot — A program that trades automatically based on rules or learned models. “Agent” usually implies more independence and adaptability than a basic bot.
Algorithmic Trading — Trading driven by computer programs following defined instructions, such as price, timing, or volume conditions.
API — Application Programming Interface. The connection that lets a trading agent talk to a broker to fetch prices and place orders.
Backtesting — Testing a trading strategy on historical data to see how it would have performed. Useful, but only reliable if done without data leakage — letting future information sneak into past decisions.
Breakout — When a price moves beyond a previous support or resistance level, often with increased volume, suggesting a new trend may be starting.
Broker — The firm that executes your trades on exchanges and holds your account.
Candlestick — A chart bar showing an asset’s open, high, low, and close prices over one period. Patterns in candlesticks are a common input for trading models.
Cloud Infrastructure — Remote servers that host trading agents, keeping them running around the clock without your own computer staying on.
Drawdown — The fall from an account’s peak value to its next lowest point. Maximum drawdown is the worst such fall in a period.
Execution — The actual process of filling an order in the market. Good execution means getting filled close to the intended price.
Kill Switch — An automatic safety control that halts trading when a limit is breached, such as a maximum daily loss.
Latency — The delay between a signal and an order reaching the market. Lower latency generally means faster, more accurate execution.
Leverage — Borrowed capital that increases both position size and risk. It magnifies gains and losses equally.
Liquidity — How easily an asset can be bought or sold without moving its price. Highly liquid markets usually mean tighter spreads and cleaner fills.
Machine Learning — A branch of AI where models improve by learning patterns from data, rather than following only hand-written rules.
Mean Reversion — The idea that prices tend to return toward their average over time. Mean-reversion strategies buy dips and sell rallies.
Model Drift — When a model’s performance degrades because market conditions change from what it was trained on.
Momentum — The tendency of assets that have been rising to keep rising, and falling assets to keep falling, for a time.
Multi-Agent System — Several specialised AI agents working together, for example one researching, one managing risk, and one executing.
Order Book — The live list of buy and sell orders for an asset. It shows where supply and demand sit at each price level.
Paper Trading — Running a strategy with simulated money to test behaviour before risking real capital.
Position Sizing — Deciding how large each trade should be, usually based on account size and risk per trade.
Prompt Injection — A security risk where malicious text in external data tricks an AI agent into following unintended instructions.
Risk Guardrail — Any automatic limit that protects capital, such as maximum drawdown caps, position limits, or kill switches.
Sharpe Ratio — A measure of return relative to risk taken. Higher generally means steadier returns for the volatility endured.
Slippage — The difference between the price you expected and the price you actually got, common in fast-moving or thin markets.
Stop-Loss — An order that automatically closes a position at a set price to limit losses.
Walk-Forward Testing — A rigorous backtesting method where a strategy is trained on one period and tested on a later, unseen period, then rolled forward repeatedly.
WebSocket — A technology for real-time data streams, letting agents receive live prices and order-book updates instantly.
Keep this AI trading glossary bookmarked. As you read deeper articles on agents, strategies, and risk, these are the terms that will keep coming up.
Further reading
This article is for education only and is not financial advice.