wdmpgWhere Do My Prompts Go?
121 sources checked · 0 fetch errors last check data v2026-10-26 · sha256 00e98854 data.jsondata.csvllms.txt

service gemini-apivendor Googlecategory Model APIs

Does Google Gemini API train on your data?

Depends (Caution)Free tier / AI Studio ('Unpaid Services'): used to improve products, and human reviewers may read it. Paid (billing-enabled project): not used. In the EEA, Switzerland and UK, paid-tier terms apply to everything.

fields
4 caution2 unfavourable1 unverified1 neutral / n/a
sources cited
4 pages
source monitor
4/4 ok on 2026-10-26
last field check

Evidence log

8 entries · newest check 2026-10-21
WARN

Trains on your data by default

= depends (Caution)

Free tier / AI Studio ('Unpaid Services'): used to improve products, and human reviewers may read it. Paid (billing-enabled project): not used. In the EEA, Switzerland and UK, paid-tier terms apply to everything.

When you use Unpaid Services, including, for example, Google AI Studio and the unpaid quota on Gemini API, Google uses the content you submit to the Services and any generated responses to provide, improve, and develop Google products

WARN

Default retention

= 'limited period' (not specified) (Caution)

Paid tier prompts are logged for abuse detection for an unspecified period. Grounding with Google Search/Maps stores data for 30 days with no way to disable it.

For Paid Services, Google logs prompts and responses for a limited period of time, solely for detecting and preventing violations of the

No fixed period stated, so this tool isn't plotted on the retention chart.

FAIL

Zero data retention

= no (use Vertex AI) (Unfavourable)

Google points customers who need guaranteed ZDR to Vertex AI instead of the Gemini Developer API.

If your workload requires guaranteed zero data retention or enterprise data processing agreements, use Vertex AI.

WARN

Abuse-monitoring exception

= always on (paid) (Caution)

Abuse logging applies to paid usage; free usage may also get human review.

To help with quality and improve our products, human reviewers may read, annotate, and process your API input and output.

N/A

Data residency / regions

= none (Neutral / n/a)

No region choice on the Gemini Developer API.

This data may be stored transiently or cached in any country in which Google or its agents maintain facilities.

UNVERIF

SOC 2 / ISO 27001

= unverified (Unverified)

No certification statement for the Gemini Developer API (AI Studio) on Google's policy pages. Google Cloud's ISO 27001 list covers Vertex AI / Gemini Enterprise Agent Platform, not this API. A Google forum reply (May 2025) says AI Studio is not ISO 27001 certified, but that is not a policy page.

no official page stated this clearly

FAIL

HIPAA BAA

= no (PHI barred without Google's written consent) (Unfavourable)

The Gemini API terms build on the Google APIs Terms, which bar HIPAA covered entities and business associates from sending PHI without Google's prior written consent. Google points regulated workloads to Vertex AI under the Google Cloud BAA; that is a separate product.

you will not use the APIs for any purpose or in any manner involving transmitting protected health information to Google unless you have received prior written consent to such use from Google.

WARN

Self-host option

= separate product (Gemini on GDC) (Caution)

The Gemini API itself is hosted only. Google sells Gemini on Google Distributed Cloud air-gapped (GA, Aug 2025) for on-premises use, which is a separate enterprise product.

We are announcing the general availability of Gemini on GDC air-gapped and preview of Gemini on GDC connected, bringing Google's most advanced models directly into your data center.

Source checks

4 pages

Spotted something wrong or out of date? Corrections are welcome; see how to send one. How statuses are assigned.

Weekly policy changes. Confirm by email. Unsubscribe anytime.

Powered by Buttondown