# Keep AI agents away from your personal data

Practice ID: 38
Topic: `ai` (AI: Boundaries that keep AI useful without surrendering sensitive context, memory, or system access.)
Added: 2026-09-01
Journey entry point: `optimize` (Make the tools you kept more durable, more private, and easier to leave.)
Start with: `technical` (technical)
Supports: `privacy-clarity` (privacy clarity; group: `digital-control` / My data and digital setup), `visible-ai-boundaries` (AI with visible, owned boundaries; group: `ai-boundaries` / My use of AI)

## Pattern

An AI that can click, run programs, or change things for you shouldn't have access to your personal apps and data by default.

## Practice

Run AI agents on a separate computer, user account, or virtual machine, or in a limited workspace, where they can't reach personal data they don't need.

## Start small

Check what one AI agent can see or run, and take away one permission it doesn't need.

## Watch out

Keeping an agent separate lowers the risk but doesn't remove it. The more it can reach, and the less you watch it, the more can go wrong.

## Related practices

- [Keep other people's data out of your AI](https://live.freewithtech.com/practices/keep-other-peoples-data-out-of-your-ai.md)

## Concepts

- [purpose-separated contexts](https://live.freewithtech.com/concepts/purpose-separated-contexts.md): Each device or software space has one clear job, so work, personal life, and other areas don't mix.

## Positions

See how these practices fit together: [Keep AI on your terms](https://live.freewithtech.com/by/keeping-ai-on-your-terms.md).

## Continue

- [Browse AI practices.](https://live.freewithtech.com/?topic=ai)
- [Journey](https://live.freewithtech.com/journey.md)
- [Concepts](https://live.freewithtech.com/concepts.md)
- [Complete agent-readable database](https://live.freewithtech.com/llms.txt)