Backfrom DigitalOcean’s own policies · verified 2026-09-02 · some facts not published
DigitalOcean: data residency & retention
Headquarters
🇺🇸 United States (US)
Where inference runs
Not disclosed
Controller: DigitalOcean, LLC (Broomfield, CO, US per privacy policy). Gradient AI docs say DigitalOcean-hosted model inference 'runs entirely within DigitalOcean's infrastructure' but do not name a country/region; third-party models (OpenAI, Anthropic) are routed to those providers.
Region pinning
Not disclosed
No region-pinning control for Gradient serverless inference is documented.
Prompt retention
Not retained
Gradient data-privacy doc: 'We do not store inputs or outputs on DigitalOcean infrastructure for any models'; sync outputs never stored, async video outputs expire after 2 hours. Exceptions via third-party models: OpenAI models through serverless inference do not support ZDR; some Anthropic models carry a mandatory 30-day trust-and-safety retention at Anthropic.
Trains on API data
No
'DigitalOcean does not use this data to train, retrain, or fine-tune any model, and does not share it with any third parties for training or fine-tuning'; data is never sent back to the original model creator.
Zero data retention
Available
Effectively ZDR by default for DigitalOcean-hosted models; not guaranteed for all third-party models (OpenAI exception documented).
Compiled from DigitalOcean’s published privacy policy, terms, and documentation on 2026-09-02. “Not disclosed” means DigitalOcean publishes nothing on the point. Policies change without notice, so verify against the sources above before relying on this for compliance decisions.