# Agentic AI for business: where it works and where not

*AI & Web3 · 7 min read · Updated 2026-10-07 · https://www.jbrichardson.com/resources/agentic-ai-for-business*

**Short answer:** An AI agent is a language model that plans and calls tools in a loop. Start with narrow, reviewable tasks such as ticket triage and drafting, give each tool least-privilege access, require human approval for any action that writes or spends, log everything, and measure success against a fixed test set.

An AI agent is a language model that works toward a goal in a loop: it decides on a step, calls a tool (search a database, create a ticket, send a draft), looks at the result and decides what to do next. That loop is powerful and, without limits, can also take wrong actions quickly.

## Good first projects

- Triage incoming support emails or tickets: classify, summarize and route, with a person approving anything customer-facing.
- Drafting replies, quotes or reports from your own data for a human to review.
- Reconciling records across systems and flagging mismatches for staff.
- Answering internal questions from your documents, with citations so answers can be checked.
- After-hours voice assistants that take messages, book appointments and hand off to a person.

## Poor first projects

- Anything irreversible, such as payments, deletions or contract changes, that runs without approval.
- Decisions about people (hiring, credit, health) where errors are costly and regulated.
- Vague goals with no way to tell whether the answer was right.

## Guardrails to build in from the start

1. Least privilege: give each tool the narrowest access it needs, read-only wherever possible.
2. Human approval for any action that writes, sends or spends.
3. Treat retrieved text (emails, web pages, documents) as untrusted. It can contain instructions meant to hijack the agent, so never let it widen the agent's permissions.
4. Log every tool call and decision so you can audit and replay what happened.
5. Set spend and step limits, and stop the loop when they are reached.
6. Keep a test set of real examples with known good answers and run it on every change.

## Measure what matters

| Measure | Why it matters |
| --- | --- |
| Task success rate on your test set | Shows whether a change helped or hurt |
| Human override rate | High overrides mean the agent is not ready for more autonomy |
| Cost and time per task | Compare against the manual process, including review time |
| Incidents and near misses | Tells you whether the guardrails are real |

> **Start small, then widen** Ship a narrow agent with approval on everything, watch it for a few weeks, and relax approvals only where the data shows it is reliable.

---
Published by JBRichardson LLC, 1603 North Olden Ave, Ewing, NJ 08638. Phone (609)-564-3016. https://www.jbrichardson.com/contact