GLOSSARY
AI Agent
Software that perceives a context, plans a course of action, calls tools, and verifies outcomes — without step-by-step human prompts. In trade, agents file customs entries, screen buyers, chase invoices, and reconcile payments.
An AI agent is software that combines a language model with the ability to call external tools, observe the results, and decide what to do next. Unlike a chatbot that answers a single question, an agent runs a multi-step workflow against a goal — for example "submit this customs entry" or "collect this invoice" — and only escalates to a human when its confidence drops below a threshold.
Why it matters
In B2B global trade, an agent can compress workflows that previously took a junior operator hours into seconds: classifying HS codes, screening sanctions lists, drafting collection emails, reconciling remittance to invoices. The accountability question — who explains the decision when CBP audits it six months later — is what makes agent observability and explainability load-bearing requirements, not nice-to-haves.
Related terms
- Model Context Protocol (MCP)
- Agent Observability
- Human-in-the-Loop
- High-risk AI System