Skip to main content
CloudLast verified June 29, 2026

xAI Grok data retention policy

xAI's API documentation states it never trains on your API inputs or outputs without explicit permission, with a 30-day audit retention window and enterprise zero data retention available. Crucially, xAI's consumer Privacy Policy does not apply to the API. Verified summary plus how to route Grok through Meetily as a BYOK summary provider.

Quick policy snapshot

Default retention
30 days (default)
Zero data retention available
Yes
Trains on API customer data by default
No

xAI's API training default

xAI's API security documentation is explicit: xAI never trains on your API inputs or outputs without your explicit permission [1]. That default applies to traffic sent to the xAI API under your own API key, which is the path Meetily uses for BYOK summaries.

This is an important distinction to draw, because xAI's consumer Privacy Policy explicitly does not apply to the API. The Privacy Policy states that it does not apply to data that xAI processes on behalf of customers of its business offerings, such as the xAI API [2][3]. Instead, API traffic is governed by xAI's Data Processing Addendum, under which xAI acts as a processor on the customer's behalf [3]. Reporting and guidance about consumer Grok on X (where public posts and Grok conversations are used for training by default unless you opt out [4]) describes the consumer surface, not the API. None of that consumer-side training behavior describes what happens to a Meetily BYOK request.

Retention

By default, xAI's documentation states that API requests and responses are temporarily stored on its servers for 30 days in case they need to be audited for potential abuse or misuse [1]. This is an abuse-monitoring retention window, not a training corpus.

For exact processing terms, geographic scope, and the current subprocessor list, the authoritative source is xAI's Data Processing Addendum, which is incorporated into the Enterprise Terms of Service [3]. Reviewing the DPA before deployment is recommended for any regulated workload.

Zero data retention

xAI offers Zero Data Retention as an enterprise feature that prevents xAI from storing any API request or response data, and it is exclusively available to enterprise accounts [1]. Under ZDR, the standard 30-day audit retention does not apply: prompts, completions, and associated metadata are processed in real time but not persisted, and once a response is delivered no record of the exchange remains [1]. Safety and content moderation checks still run in real time, but their results are not stored under ZDR [1]. To enable it, contact xAI sales; it is not a self-serve toggle for standard accounts.

How Meetily uses xAI Grok

Meetily transcription is always 100% local, and audio never leaves your device. When you select an xAI Grok model as your summary provider in the BYOK flow, Meetily sends transcript text to the xAI API over TLS using your own API key, and the summary is returned to Meetily and stored locally on your device.

The xAI API is OpenAI-compatible: xAI documents that you can use the OpenAI SDK by pointing it at the base URL https://api.x.ai/v1, and most OpenAI-compatible tools work without modification [5]. Meetily configures it as an OpenAI-compatible base URL, so your traffic runs under your own xAI account and its contract terms, not Meetily's.

If you want stronger guarantees than the cloud default offers, switch your summary model to a local provider such as Ollama. Meetily's transcription path is already local, so the local-summary route keeps the entire pipeline on-device.

References

  1. FAQ - xAI API Security, xAI (publisher: xAI). https://docs.x.ai/developers/faq/security. Accessed 2026-06-29.
  2. Privacy Policy, xAI (publisher: xAI). https://x.ai/legal/privacy-policy. Accessed 2026-06-29.
  3. Data Processing Addendum, xAI (publisher: xAI). https://x.ai/legal/data-processing-addendum. Accessed 2026-06-29.
  4. About Grok, X Help Center (publisher: X Corp.). https://help.x.com/en/using-x/about-grok. Accessed 2026-06-29.
  5. API: Frontier Models for Reasoning & Enterprise, xAI (publisher: xAI). https://x.ai/api. Accessed 2026-06-29.

Last verified: June 29, 2026. Policy source: xAI Grok policy

Frequently asked questions

Does xAI train Grok models on my API prompts and outputs?
No, not by default. xAI's API security documentation states that xAI never trains on your API inputs or outputs without your explicit permission. This page covers the xAI API path only. Note that this is a different posture from consumer Grok on X, where public posts and Grok conversations are used for training by default unless you opt out in X settings.
Why does this page say the API is different from xAI's consumer Privacy Policy?
Because xAI says so explicitly. xAI's Privacy Policy states it does not apply to data that xAI processes on behalf of customers of its business offerings, such as the xAI API. API traffic is instead governed by xAI's Data Processing Addendum, under which xAI acts as a processor on your behalf. So any training or retention behavior described in the consumer policy or in coverage of Grok-on-X does not describe what happens to your API requests through Meetily BYOK.
How long does xAI retain my API prompts and outputs?
By default, xAI's documentation states that API requests and responses are temporarily stored for 30 days in case they need to be audited for potential abuse or misuse. With Zero Data Retention enabled (enterprise accounts), that 30-day audit retention does not apply and request and response data is not persisted.
Is zero data retention available for the xAI API?
Yes, for enterprise accounts. xAI describes Zero Data Retention as an enterprise feature that prevents xAI from storing any API request or response data, and notes it is exclusively available to enterprise accounts. Safety and moderation checks still run in real time, but under ZDR the results are not stored. Contact xAI sales to enable it; it is not a self-serve toggle on standard accounts.
Where is xAI API data processed geographically?
xAI processes API personal data as a processor under its Data Processing Addendum. The default endpoint is US-based, and xAI also publishes regional endpoints (for example a Europe endpoint) so requests can be routed to a chosen region. Exact data-residency commitments and the current subprocessor list are governed by the DPA referenced from the Enterprise Terms; review the current DPA for your jurisdiction before deploying a regulated workload.
How does Meetily handle Grok when I pick it as my summary provider?
Meetily transcription is always 100% local. When you select an xAI Grok model as your summary provider in the BYOK flow, Meetily sends transcript text (not audio) to the xAI API using your own API key. The xAI API is OpenAI-compatible, so it is configured as an OpenAI-compatible base URL (https://api.x.ai/v1). The retention and training defaults on this page apply to your own API traffic, not to Meetily.
What if I want zero retention but my account does not have ZDR enabled?
Two options: ask xAI sales to enable Zero Data Retention on an enterprise account, or switch your Meetily summary model to a local provider like Ollama. Meetily's transcription path is already local, so the local-summary route keeps the entire pipeline on your device with nothing sent to any cloud provider.

Use xAI Grok with Meetily, on your terms

Meetily transcription stays 100% local. For summaries, bring your own xAI Grok key (BYOK) so the data path matches the policy you just read - or pick a local model if you want zero retention by construction.

Set up BYOKDownload Meetily