Legal

Privacy Policy

Last updated: May 17, 2026

Drift is a desktop IDE that runs on your machine. We designed it so that your code, conversations, and credentials stay under your control. This page explains what data Drift handles, where it lives, and the limited cases in which it leaves your device.

Summary

  • Your code never reaches Drift servers. When you use AI features, requests go directly from your machine to the model provider you configured (Anthropic, OpenAI, Google).
  • You bring your own API keys. Credentials are stored locally — in your operating system keychain or in the provider's own credential file — and are never transmitted to us.
  • Telemetry is anonymous and aggregate. We collect basic usage events to improve the product. No code, prompts, file contents, or personally identifying data are included.
  • A Drift Cloud account is optional. It is only required for features that explicitly opt in, such as account-based settings sync or paid subscriptions.

What stays on your device

The following data is stored locally and never transmitted to Drift unless you explicitly choose to share it (for example, by attaching a log to a bug report):

  • Source code, projects, and workspace files
  • Chat conversations and session history with AI agents
  • API keys and OAuth credentials for Claude, Codex, Gemini, and other providers
  • Editor settings, keybindings, extensions, profiles, and themes
  • SSH connection details and remote workspace state
  • MCP server configurations and custom skills

What is sent to AI providers

When you send a message to an agent, Drift forwards your prompt and the relevant context (file excerpts, command output, tool results) directly to the provider you selected. Drift does not proxy these requests through its own infrastructure. Each provider's privacy and data-retention policies apply:

You choose which provider to use for each session. You can revoke a provider's access at any time by removing the corresponding API key or signing out of its OAuth integration from within Drift.

Telemetry

Drift collects anonymous usage telemetry to understand which features are used, diagnose crashes, and improve performance. Events include things like which view was opened, which model was selected, and whether a session completed successfully.

Telemetry never includes the contents of your prompts, files, terminal output, chat history, file paths outside of the application itself, or any personally identifying information. Events are associated with a randomly generated anonymous client identifier, not with your account or device.

You can disable telemetry at any time by setting telemetry.telemetryLevel to off in Drift's settings. The setting takes effect immediately and applies to every subsequent session.

Drift Cloud account (optional)

You can use Drift entirely offline without an account. If you choose to create a Drift Cloud account (at account.drift-ide.com), we collect the information needed to operate the account service:

  • Email address and authentication credentials
  • Subscription, billing, and entitlement records (for paid plans)
  • Session tokens used to authenticate the IDE to the account service

Account data is stored on infrastructure we operate and is used only to provide the services you signed up for. We do not sell account data, and we do not use it to train AI models.

Remote SSH

Drift includes a built-in SSH client that connects your IDE to remote machines you control. SSH connections are established directly between your computer and the target host. Credentials, host keys, and remote file contents are not transmitted to Drift.

Updates and crash reports

The application periodically checks for new releases by contacting our update server. Crash reports, when enabled, include a stack trace and the runtime environment (operating system version, application version). They do not include source code or chat history.

Third-party services

Some Drift features connect to third-party services when you use them — for example, MCP servers you configure, web pages an agent fetches on your request, or extensions installed from the marketplace. These interactions are governed by those services' own privacy policies. Drift does not collect or store data on your behalf for these interactions.

Children

Drift is not directed at children under 13, and we do not knowingly collect personal information from them.

Changes to this policy

We may update this policy as the product evolves. Material changes will be announced through the application or via email if you have a Drift Cloud account. The "Last updated" date at the top of this page reflects the most recent revision.

Contact

Questions about this policy or a request related to your data can be sent to [email protected].