# Edge AI

> AI that runs on the device itself. Think of a phone, a camera or a car.

**Field:** Artificial Intelligence  
**Category:** Running Locally  
**Also known as:** Edge Inference, AI at the Edge, Edge Machine Learning

## Think of it like

It is like a guard at the door who decides on the spot. There is no call to the boss.

## Definition

Edge AI is AI that runs on the device where the data is made. That could be a phone, a camera, a car or a sensor in a factory. The device makes its choice right there, with no trip to a far server. That makes it fast, and it keeps working with no internet. The data can stay on the device too. The models are small, because the chips in these devices are small.

## Example

A doorbell camera spots a person at the door. A tiny model inside the camera makes that call in a split second. Only the alert goes out to the owner's phone.

## Why it matters

Some choices cannot wait for the internet. A car that has to brake needs the answer now.

## Common mistake

Thinking edge AI means a big model on a strong PC. It most often means a small model, shrunk to fit a small device.

## Three principles

- **On the device.** No trip to a far server.
- **Fast and offline.** It works with no signal.
- **Small models.** Built to fit small chips.

## Related terms

- [Local AI Model](https://learn.benibix.com/term/ai/local-ai-model)
- [Small Language Model (SLM)](https://learn.benibix.com/term/ai/small-language-model)
- [Quantization (AI)](https://learn.benibix.com/term/ai/quantization)
- [Model Pruning](https://learn.benibix.com/term/ai/model-pruning)
- [Latency](https://learn.benibix.com/term/infrastructure/latency)

![Two paths for the same camera picture. On the left, the camera sends the picture over the internet to a far server, which decides and sends the answer back. On the right, highlighted, a small model inside the camera decides on the spot, and only a short alert leaves the device.](https://learn.benibix.com/diagrams/ai4-edge-ai.svg)

---

Source: https://learn.benibix.com/term/ai/edge-ai
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-19
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
