NEWS
Verax AI Risk Assessment is live. See what's exposed
Cursor
Cursor
 
Risk Overview
Moderate Risk
4.6
/10
Code Generation
·
Code Completion
·
Top Risks

Mandatory arbitration + class action waiver. Default training on potentially proprietary code. Very high agentic risk — reads entire codebase, terminal access, file system.

Mitigations

ENABLE PRIVACY MODE immediately. Enterprise plan enables Privacy Mode by default. Opt out of telemetry.

Get started
Is your team using 
Cursor
 
safely?

Highest Risk Categories

The highest-scoring risk categories identified in this assessment.

Legal Terms

Mandatory arbitration (JAMS). Class action waiver. California law. Very similar to OpenAI legal posture.
8
/10

Training Use

Default: trains on usage. Privacy Mode disables training entirely. Enterprise: Privacy Mode on by default.
6
/10

3rd-Party Sharing

Anthropic/OpenAI as LLM providers with ZDR when Privacy Mode active. No ad tracking. Standard service provider flows.
4
/10

Unlock the complete assessment

Run a free assessment to access the full AI tool risk analysis.
Start Free Assessment
Updated July 2026 • AI-assisted research • Reviewed by Verax • Report an issue

Reference Documentation

Official documentation and policies referenced as part of this assessment.