---
title: "Sub-processors | Minds"
canonical_url: "https://getminds.ai/legal/subprocessors"
last_updated: 2026-09-25
meta:
  description: "Last Updated: September 25, 2026"
  "og:description": "Last Updated: September 25, 2026"
  "og:title": "Sub-processors | Minds"
  "twitter:description": "Last Updated: September 25, 2026"
  "twitter:title": "Sub-processors | Minds"
---

Minds

September 25, 2026·Minds Team # **Sub-processors** Last Updated: September 25, 2026**Last Updated: September 25, 2026** Art of X UG (haftungsbeschränkt) uses the following sub-processors to provide our services. This list is updated when changes occur. ## Current Sub-processors | Sub-processor | Purpose | Location |
| --- | --- | --- | | **Art of X UG (haftungsbeschränkt)** | Platform development, hosting, operation, security, and support. Applies where Minds AI Labs, Inc. is the contracting entity; every other sub-processor on this list then acts beneath it. | Germany (Berlin) | | **Cloudflare** | CDN, DNS, DDoS protection, web security | USA (global edge network) | | **DigitalOcean** | Cloud infrastructure, hosting | Germany (Frankfurt) | | **Supabase** | Database, authentication | EU (Stockholm) | | **OpenAI** | LLM API for AI processing, embeddings, transcription | USA | | **Anthropic** | LLM API for AI processing | USA | | **Google Cloud Vertex AI** | LLM API for AI processing and AI image generation, accessed through Vertex AI on Google Cloud | USA | | **Vercel (Vercel Inc.)** | AI inference routing gateway (Vercel AI Gateway): forwards OpenAI and Anthropic model requests, including OpenAI embedding requests, to those providers' first-party business APIs; also forwards evaluation requests to TypeSafe AI, namely requests containing only public web source text (evidence screening) and requests containing only AI-generated Study answer text (language screening) | USA (primary) | | **TypeSafe AI (TypeSafe AI, Inc.)** | Evaluation model (Jev) used for two screening tasks: (1) evidence screening, which screens public web source text for statistical data during AI audience research and receives public web text only, no user data; (2) language screening of Study answers, which checks whether an answer generated by a Mind is written in the Study's requested response language and receives only that AI-generated answer text (which may relate to the customer's Study content) and the language name; accessed only through Vercel AI Gateway; zero data retention requested | USA | | **Exa (Exa Labs)** | Web search, public-page content retrieval, and AI audience-research grounding; queries may include submitted URLs, audience descriptions, or persona subject names | USA | | **Firecrawl (SideGuide Technologies)** | Web content extraction, visual page analysis | USA | | **Browserbase** | Managed browser infrastructure for user-requested website analysis | USA | | **Langfuse** | LLM observability and tracing | Germany | | **Sentry (Functional Software, Inc.)** | Application error monitoring and performance tracing | Germany (EU region) | | **Resend** | Transactional email delivery | USA | | **Stripe** | Payment processing | Ireland (EU data) | | **PostHog** | Product analytics | EU | | **Google Analytics** | Web analytics | USA | | **TikTok** | Conversion tracking via TikTok Pixel (marketing consent required) | Ireland/USA | | **X (Twitter)** | Conversion tracking via X Pixel (marketing consent required) | USA | | **Meta Platforms (Lead Ads)** | Receiving contact details submitted through our Meta lead ad forms | USA / EU (Ireland) | | **Twilio** | SMS, voice calls, phone number provisioning | USA | | **Deepgram** | Speech-to-text transcription | USA | | **Apify** | Web content extraction, social media scraping | Czech Republic (EU) | | **RevenueCat** | In-app purchase management (iOS) | USA | | **Slack (Salesforce)** | Internal notifications (webhooks) | USA | | **Google Workspace (Gmail)** | Business email: sending team correspondence incl. product follow-up email, and syncing replies to the CRM | Ireland/USA | | **Google (YouTube Data API)** | Discovery of publicly available videos for AI persona knowledge (search queries may include persona subject names) | USA | | **Google Drive API** | File attachments from Google Drive (user-selected files) | USA | | **Microsoft Graph API (OneDrive)** | File attachments from OneDrive (user-selected files) | USA | | **Figma API** | Design attachments from Figma (user-selected files, rendered to images) | USA | | **Linear API** | Issue and PRD imports from Linear (user-selected issues) | USA | | **Atlassian (Jira API)** | Issue and PRD imports from Jira (user-selected issues) | USA | | **Meta Platforms (Marketing API)** | User-authorized, read-only imports of ad copy and creative assets | USA | ## User-selected external clients External clients chosen by the user to connect to Minds (for example ChatGPT, Claude, GitHub Copilot, Codex, Gemini CLI, Le Chat, Manus, Windsurf, Cursor, OpenRouter, Open WebUI, LibreChat, or similar tools) are independent services. They are not Art of X sub-processors unless Art of X engages them to process data on its behalf. Their processing is governed by their own terms and privacy notices. ## AI Inference Routing AI inference is carried out through the providers' business interfaces, not through consumer products: - **OpenAI** and **Anthropic** models are accessed through the providers' first-party business APIs, either directly or through **Vercel AI Gateway** (Vercel Inc.). When the gateway is used, it forwards each request to the model creator's own first-party business API: the Anthropic API for Claude models and the OpenAI API for GPT models and OpenAI embeddings. The gateway is restricted to these first-party backends and does not route our requests to other hosting providers of the same models. Other OpenAI features, such as speech transcription, are accessed directly through OpenAI's API. - **TypeSafe AI**'s evaluation model **Jev** is accessed only through **Vercel AI Gateway**. Each request is restricted to TypeSafe AI as the only provider and is sent with the gateway's zero-data-retention option enabled. Evidence-screening requests contain only text from public web sources (web pages, PDFs, and tables retrieved from the open web) and a fixed screening question; no audience brief, uploaded files, or account, persona, or respondent data are sent. Language-screening requests contain only the text of an answer generated by a Mind in a Study (which may relate to the customer's Study content), the name of the Study's requested response language, and a fixed screening question; the Study question, uploaded files or attachments, source material, audience briefs, and account, persona, or respondent profile data are not sent. - **Google** models are accessed exclusively through **Vertex AI on Google Cloud**. The consumer Gemini API is not used. Where requests are routed through Vercel AI Gateway, Vercel acts as our sub-processor and engages OpenAI, Anthropic, or TypeSafe AI to process them. Each of them remains listed above, and the requirements in the Data Transfer Safeguards section apply to each processor in that chain. Under the applicable business terms, content submitted for inference is not used by these providers to train their generally available models. According to Vercel, the gateway does not use prompts or outputs for model training and does not retain them after a request is completed. This applies to the gateway, not to the model providers. According to TypeSafe AI's privacy policy, TypeSafe AI does not train or fine-tune AI or machine-learning models on customer prompts or other input. Providers may retain limited copies for a short period for abuse monitoring and security, unless a separate zero-retention arrangement applies. Where we state that data is not used for training, that statement is not the same as, and should not be read as, a zero-data-retention guarantee. ### Processing region and customer-provided endpoints Vertex AI processing regions are configurable, including European regions. Where a customer requires inference to be processed in a specific region, or wishes to route inference to its own, self-hosted, or separately contracted model endpoint, this is arranged individually as part of an enterprise agreement rather than selected in the product interface. The applicable region and endpoints are then recorded in that agreement. Contact [privacy@getminds.ai](https://getminds.ai/mailto:privacy@getminds.ai) to scope this before relying on it for a transfer assessment. ## Data Transfer Safeguards For sub-processors located outside the European Economic Area (EEA), the applicable transfer mechanism and contractual measures are assessed for each vendor and processing activity. The legal requirement for processor terms does not depend on whether a vendor makes its preferred contract available. Depending on the transfer, safeguards may include: - **Standard Contractual Clauses (SCCs)** where required - **EU-US Data Privacy Framework** certification where applicable - **Data Processing Agreements or equivalent contractual terms** where legally required If legally required processor terms or transfer safeguards cannot be put in place, the affected processing must not begin or must be suspended until the gap is resolved. Additional technical and organizational measures, including data minimization, may be required by the assessment. Approved sub-processors may process Customer Data only for the documented service purpose and are subject to applicable confidentiality, security, retention, deletion, transfer, and no-generalized-training requirements. Customer-selected independent services remain governed by the separate notice above. ## Changes to Sub-processors We will notify customers of any intended changes to sub-processors with at least 14 days notice before the change takes effect. Notifications are sent to the email address associated with your account. To object to a new sub-processor, contact us at [privacy@getminds.ai](https://getminds.ai/mailto:privacy@getminds.ai) within 14 days of notification. ## Questions For questions about our sub-processors, contact: [privacy@getminds.ai](https://getminds.ai/mailto:privacy@getminds.ai) ---**Art of X UG (haftungsbeschränkt)** Köpenicker Straße 145, 10997 Berlin, Germany ## **Continue your procurement review** Use the buyer-facing checklist and evidence pages alongside these legal terms. [Minds](https://getminds.ai/)© 2026 Minds. Your target audience. AI-driven and grounded in transparent evidence. Build within minutes. [Minds on X (Twitter)](https://x.com/mindsai_co) [Minds on LinkedIn](https://www.linkedin.com/company/mindsaicompany/) [Minds on Instagram](https://www.instagram.com/getminds.ai/)Minds is part of [![ESOMAR](https://getminds.ai/images/newsroom/logos/esomar-logo.svg)ESOMAR](https://esomar.org/) [![GreenBook](https://getminds.ai/images/newsroom/logos/greenbook.svg)GreenBook Directory](https://greenbook.org/company/Minds) [![Insight Platforms](https://getminds.ai/images/newsroom/logos/insight-platforms.png)Insight Platforms](https://www.insightplatforms.com/platforms/minds/) [![Capterra](https://getminds.ai/images/newsroom/logos/capterra.svg)Capterra](https://www.capterra.com/p/10046203/Minds/) [![G2](https://getminds.ai/images/newsroom/logos/g2.svg)G2](https://www.g2.com/products/minds/reviews) [![CSSDA Best UX Design Award](https://getminds.ai/images/newsroom/logos/cssda-best-ux-award.png)CSSDA Best UX Design Award](https://www.cssdesignawards.com/) [![CSSDA Best Innovation Award](https://getminds.ai/images/newsroom/logos/cssda-best-innovation-award.png)CSSDA Best Innovation Award](https://www.cssdesignawards.com/) [![CSSDA Best UI Design Award](https://getminds.ai/images/newsroom/logos/cssda-best-ui-award.png)CSSDA Best UI Design Award](https://www.cssdesignawards.com/)