AI Use & Data Policy

What AI features Forge uses

• Issue triage — AI-generated severity suggestions and duplicate detection when a bug is filed.
• Think Tank sounding board — AI lenses that challenge or expand on ideas you submit.
• Standup & AI digest — Summaries of open issues and sprint progress sent via email or Slack.
• PR impact prediction — Risk analysis for linked pull requests.
• Release notes — AI-generated summaries of completed issues for a date range.
• Support triage — Initial guidance on incoming support tickets.

Which AI providers process your data

By default, Forge routes AI requests to xAI (Grok). Workspace admins can supply their own API keys for OpenAI, Anthropic (Claude), Google Gemini, or xAI — in that case, requests go directly to the provider you configure. See our sub-processor list for details.

How your data is used

• Issue text, comments, and project names may be sent to your configured AI provider to generate responses.
• No personal financial, health, or government ID data should be entered into Forge issues or comments.
• Your data is never used to train AI models. We use inference-only API endpoints with zero data retention commitments from providers.
• AI inputs and outputs are logged in idea_ai_turns / audit_log for your own auditability. You can request deletion of this data via a GDPR/CCPA data deletion request.

Data retention

AI interaction logs are retained for as long as your workspace is active. If you disable AI features in Admin → Settings → AI, no new AI requests are sent. Workspace owners can delete all AI interaction logs by submitting a data deletion request via Admin → Compliance.

Opting out

Workspace owners and admins can disable AI features globally in Admin → Settings → AI. Individual users cannot opt out of workspace-level AI triage, but can opt out of personal email digests in Account → Preferences.

Contact

Questions about AI data handling? Email privacy@forge.app.